DYNAMICS OF
ORGANIZATIONAL LEARNING
Promotor
:
prof.dr. A.A.I. Holtgrefe
Copromotor
:
dr. M.S.H. Heng
Referent
:
prof. J.G. March
TO MY PARENTS
PREFACE
This book reflects a doctoral research conducted from April 1992 to October 1996.
In this period, I tried to explore the concept of organizational learning with the idea of
providing it with more theoretical underpinnings than it had known in the past.
Although the concept of organizational learning is very popular at the moment, it
lacks any mutually agreed upon description, definition, or theory. In fact, the interest in
organizational learning is so widespread that its literature reflects at least six different
perspectives. In this thesis I have tried to integrate these perspectives by borrowing the
most valuable aspects of each one and by trying to avoid various biases also present in
them. In addition, partly as a result of this integration, I have tried to introduce a more
'realistic' perspective on learning. In this preface, I will elaborate a bit more on the latter
contribution of this thesis.
One of the most popular perspectives in the literature at the moment is the one
called the "Learning Organization." A learning organization can be seen as a specific type
of organization which is designed to foster positive learning outcomes such as
improvement, intelligence, and innovation. These outcomes can be realized through
flexible work design, teams, open communication, and an inspiring and supportive style of
management. It is very clear that these and other ingredients of a learning organization do
sound attractive. However, the problem is that current researchers cannot produce
convincing arguments about the dynamics which underlie the process of learning. In fact,
the literature remains conspicuously silent on the question of how such organizational
delight is to be achieved.
In order to contribute to closing this knowledge gap, I choose to approach
organizational learning as a process instead of as an outcome. A process perspective is
more likely to reveal the underlying dynamics of learning which either produce or impede
positively valued outcomes. In fact, every organization is a learning organization whose
nature depends basically on the underlying dynamics of learning.
I perceive organizational learning as a process of evolution: the organization
develops over time as a result of construction and reconstruction of organizational
knowledge. This development can result in improvement, but it can also cause inertia or
even self-destruction.
Organizations learn through their members: organizational members introduce new
knowledge into the organization which is subsequently shared by other members. This is
how it becomes organizational knowledge. Organizational knowledge, in turn, influences
subsequent processes of introducing knowledge. These processes of externalization,
objectification, and internalization of knowledge, is what I term "Internal Learning".
Internal learning can be susceptible to various inefficiencies. For example, determination
of what knowledge becomes organizational knowledge is often influenced by the exercise
of power.
There are various ways in which new knowledge is introduced into the
organization. To give an example, the university as an organization may learn from its
students - through for instance participation in committees or during day to day encounters
- that its current activities are not meeting student needs. When this feedback knowledge
results in changing the curriculum, in changing the ways it approaches students, or in
changing perceptions of its relation with the environment, then the university has learned.
This so-called "Feedback learning" can be influenced by various underlying dynamics
which may in fact complicate the learning process. For example, the university may
translate feedback information it receives in a different way from the one students
intended. Or, it may select only certain environments from which it wants to learn, and
filter out or exclude others.
New knowledge can also be introduced through the imitation of other
organizations. Take for example an organization that hires a business consultant in the
hope of solving some persistent organizational problems. This consultant brings with her
specific knowledge that she has gained during her professional education as well as during
her previous consulting projects. Say that this knowledge concerns a new system design
methodology. After a period of time in which the methodology comes to be implemented,
organizational members get used to this new system design; the new methodology
becomes part of organizational knowledge and is then taken for granted. At this point, we
could say that the organization has learned a new methodology through diffusion of
external knowledge. This particular learning process is called "Learning from others".
However, learning from others may also result in various inefficiencies if, for example,
existing organizational knowledge does not match the new knowledge, or if the
organization has only limited access to models it can potentially imitate.
Finally, introduction of new knowledge can be a result of experimenting and
creativity. Instead of adapting to external knowledge, the organization learns from new
ideas created by its members. Imagine an organizational member - motivated to alleviate
some anxiety in his work - who starts thinking about possible alternative courses the
organization can take. While discussing present use of an existing technology, he
stimulates a group of people to explore possible alternative usages. Through brainstorming
and informal networking, new ideas are generated, tried out, adjusted, and introduced to a
larger group of organizational members. Over time, the new idea results in an innovation:
the organization uses the technology for any purpose it deems fit.
This type of success story resembles learning which occurs in so-called "Learning
Organizations." However, there are various tendencies which obstruct this "Creative
learning" process. For example, organizations are often risk averse; this means that there
is a shortage of slack resources in terms of money, experience, skills, and external
contacts at their disposal. Conversely, organizations can also be too risk seeking and
thereby unable to exploit experiences gained during experimenting.
Avoiding inefficiencies which may occur during various types of learning, is
clearly one step toward successful outcomes of learning. What is also needed is to balance
the four types of learning. If one type of learning predominates in the organization,
negative outcomes may be produced, either in the short or the long run. For instance,
when an organization relies mainly on feedback information, its evolution will most likely
be characterized by path dependency. It learns from its own experience without
considering alternative courses of action. To avoid this path dependency, organizations
need to complement their learning with the experience of other organizations as well as
with experience obtained through creativity and experimenting. The same is true for other
types of leaning: too much focus on one type of learning results in path dependent
evolution or other negative outcomes of learning.
Because of these and many other dynamics that underlie organizational learning, a
more realistic view is needed to approach the concept than has been previously put
forward. Although it is very attractive to imagine "learning organizations" which are
flexible, innovative, tolerant, and inspiring, it is first of all necessary to analyze the way
organizations really learn, to understand the complexities which may frustrate positive
outcomes from occurring, and to explore possibilities for coping with these complexities.
If such a process is realized, then more successful outcomes of learning might be
promoted.
In the years I conducted my research, I encountered various reactions from at least
four groups of individuals.
One group asked me whenever I told them about my research: "What is
organizational learning anyway?". This group mainly consisted of people I knew from
outside academic and organizational life. In pursuing this matter, they often gave me a
hard time, especially at the beginning of my research. Depending on the time and on the
context, I gave them different answers. To be honest, it took me years to understand what
organizational learning could mean.
I hope that these people will read what I have written, or at least ask me the
question once again.
A second group reacted to my endeavors by thinking or saying: "Let's see what
she makes of it all...". This group consisted mainly of fellow researchers who had also
studied organizational learning; some had just started their projects, others began research
but eventually gave up, or replaced the word 'learning' with 'change.'
I am curious to know what they will think of this thesis. Although the work had to
be finished because of time constraints, it is still far from complete. I am looking forward
to continuing our discussions about the theoretical foundations of 'organizational learning'.
A third group of individuals used to think - although they often refrained from
saying -"Oh no, not again!" In general, this group consisted of organizational practitioners
who became fed up with all the books, conferences, and workshops which dealt with this
subject. These people experienced all the output, but still did not have a clue what
organizational learning meant.
Clearly, this is not due to their own ignorance; rather the opposite. So much of
what has been said about organizational learning (or the learning company) in the past few
years may appear attractive at first glance. However, upon further reflection, the texts
ring hollow as the reader takes the time to think them over. I hope my fellow enthusiasts
do not give up completely, and are still willing to read this thesis on organizational
learning.
Fortunately, there was still a fourth group. These people exclaimed "Interesting!"
whenever I related my research efforts. Either they did not give up on the project, or they
sought to promote learning processes in their own organizations, or, they were just nice
people.
I hope they will enjoy reading the book.
Acknowledgements
I would like to thank everyone who helped me think, search, and explore, and who
provided an inspiring and supportive environment.
First of all, there is Michael Heng. He stimulated me to think big and act small: an
attractive philosophy since it not only inspired me to read classics and to explore the
fundamentals of organization theory but it also kept me active and optimistic during the
whole project. I thank him for initiating the research and for his advice throughout the
four years of my study.
Some students require strict guidance from their advisors, especially at the
beginning of their research. Others are happy in a more playful environment. I certainly
belong to the latter group. I am very grateful for Guus Holtgrefe, my promotor, for
acknowledging this. I also thank him for reviewing the various drafts of the thesis.
Of course, starting research by playing around has the danger of letting the student
sink into academic quicksand. In addition to Michael and Guus, there are many other
people who helped prevent this from happening to me. In the early beginnings there were
Martin Boogaard and Sven Fischer who showed me that being a Ph.D. student and writing
scientific papers is not such a tedious job after all. I also profited a lot from discussions
with Jan Achterberg, Rob Buitendijk, Herman Coenen, Rob Hundman, Steve Peters, and
Anselm Strauss, who helped me understand what I was doing and gave me the opportunity
to carry out the research.
Then of course, there are my other colleagues with whom I like to work, to
discuss research, to talk babies, to drink beer, and to have fun: Heico van der Blonk,
Marcel Creemers, Frank Derksen, Han Gerrits, Ypke Hiemstra, Astrid Mantel, Irene
Lammers, Martin Misseyer, Mike Newman, Gerlof Sijtsma, Edu Spoor, René Wagenaar,
and Marcel Wieser. Together they make the "BIK" group a very pleasant environment.
A welcome change in research life was my stay at Stanford University from
December 1993 until September 1994. In these months, I learned that reading and
conducting organizational research can be very exciting. I owe a lot for that to Jim
March. He showed me the fun side of organizational life, both theoretically and in
practice. Above all, I have profited a lot from his advice on my research and his
introduction to the 'basics' of organizational learning.
I also have good memories of the many inspiring discussions I had with the
researchers I met at Stanford. In particular I thank Kjersti Halvorsen and Philipp Genschel
for cheering up Palo Alto. Many thanks also to Barbara Beuche for her ever-lasting
friendly assistance and care.
Back in Holland, it was time to stop playing around and actually set down in words
what I had learned. A period of solitary research set in. Fortunately, in this period I met
Ariane von Raesfeld who prevented me from becoming completely isolated. I really
enjoyed the challenging 'interactions' we had both socially and scientifically.
After the first 'final' version of the thesis was completed, I asked some people to
read it. I am very grateful to both Ariane von Raesfeld and Heico van der Blonk for their
serious review of the thesis. Harriet Mowshowitz was the last one who helped me with the
completion of the work. I very much appreciate her help in editing the text.
Back home, there are my family and friends to whom I am grateful for showing
me that there is more to life than doing research.
Some people think that combining research with a baby is a very strenuous job;
and they are probably right. In my case I was lucky. Thanks to Abel for being so sweet,
for keeping me happy while I struggled with the thesis, and for making my life a bit more
orderly. Of course, I owe a lot to my Leo who carried me - as always - through the whole
project. Finally, I thank my parents for all their support and enthusiasm. To them I
dedicate this book.
TABLE OF CONTENTS
CHAPTER 1
INTRODUCTION
1.1
Introduction
1
1.2
Foundations of organizational learning
4
1.3
Research methodology
10
1.4
Overview of chapters
16
PART ONE
IN SEARCH OF A THEORY OF ORGANIZATIONAL LEARNING
CHAPTER 2
DIFFERENT PERSPECTIVES ON ORGANIZATIONAL LEARNING
2.1
Introduction
21
2.2
Adaptation perspective
21
2.3
Incremental innovation
24
2.4
Assumption sharing perspective
26
2.5
Organizational knowledge perspective
28
2.6
The 'learning organization'
30
2.7
Social constructivist perspective
32
2.8
Concluding remarks
35
CHAPTER 3
A CRITICAL REVIEW
3.1
Introduction
37
3.2
Improvement bias
38
3.3
Individual action bias
41
3.4
Systems theory bias
44
3.5
Planned and strategic learning bias
46
3.6
One or two-sided learning bias
48
3.7
Concluding remarks
53
PART TWO
A TYPOLOGY OF ORGANIZATIONAL LEARNING PROCESSES
CHAPTER 4
INTERNAL LEARNING
4.1
Introduction
59
4.2
The concept of organizational knowledge
59
4.3
The process of institutionalizing knowledge
61
4.3.1 Externalization of individual knowledge
64
4.3.2 Objectivation of knowledge
66
4.3.3 Internalization of knowledge
67
Intermezzo I "Learning from newcomers at AZ"
70
4.4
Traps and obstacles during internal learning
77
4.5
Concluding remarks
83
CHAPTER 5
FEEDBACK LEARNING
5.1
Introduction
85
5.2
Organizations and their environments
85
5.3
Theories of feedback learning
88
5.4
Traps and obstacles during feedback learning
91
5.5
Concluding remarks
94
CHAPTER 6
LEARNING FROM OTHERS
6.1
Introduction
97
6.2
Diffusion of external knowledge
97
6.3
Adopting external knowledge
101
6.4
Traps and obstacles during learning from others
104
6.5
Concluding remarks
106
CHAPTER 7
CREATIVE LEARNING
7.1
Introduction
109
Intermezzo II "Creating the Mobility Pass at LeaseCo"
111
7.2
Internally triggered variation
115
7.2.1 Enactment processes
116
7.2.2 Technology of foolishness
118
7.2.3 Diversity
120
7.2.4 Informal networking
123
7.2.5 Risk-taking
125
7.3
Traps and obstacles during creative learning
127
7.4
Concluding remarks
128
PART THREE
IMPLICATIONS
CHAPTER 8
STRIVING FOR SUCCESSFUL LEARNING
8.1
Introduction
133
8.2
Focussed learning
135
8.2.1 Specialization
136
8.2.2 Physical and cultural conditions
136
8.2.3 Reference groups
137
8.2.4 Self-reference
139
8.2.5 Hidden learning
141
8.2.6 Avoiding focussed learning
142
8.3
Unbalanced learning
147
8.4
Concluding remarks
156
CHAPTER 9
INFORMATION SYSTEMS AND ORGANIZATIONAL LEARNING
9.1
Introduction
159
9.2
Research on organizational learning and information systems
160
9.3
Externalization of knowledge
163
9.4
Objectivation of knowledge
168
9.5
Internalization of knowledge
170
9.6
Information selection
171
9.7
Information interpretation
174
9.8
Idea generation
176
9.9
Concluding remarks
181
CHAPTER 10
CONCLUDING REMARKS
10.1
Introduction
183
10.2
Summerizing the arguments so far
183
10.3
Post-modernism and organizational learning
188
10.4
Contribution to existing knowledge
191
10.5
Recommendations for further research
194
APPENDIX I (Assessing organizational learning capacity)
199
REFERENCES
205
NEDERLANDSE SAMENVATTING (SUMMARY IN DUTCH)
223
CHAPTER ONE
INTRODUCTION
1.1
INTRODUCTION
The concept of organizational learning has never been as popular as it is today.
The odds are high that random browsing through one of the many organization and
management journals will reveal at least one article on the subject. Organizational learning
seems to be a bandwagon every organization wants to get on or should at least try to get
on. Strange as it may seem, there is at the same time a lack of a shared understanding of
its meaning.
What is organizational learning? Ask a random researcher or practitioner within the
world of organizations, and again the odds are high that he or she cannot give you a
satisfactory answer. General awareness about the importance of organizational learning
seems to be combined with a general ignorance about its meaning. Although articles,
1
books, congresses, special issues of journals on the topic proliferate , consensus about
what learning is and how it occurs is difficult to achieve. Multiple perspectives exist at the
same time, each isolating interesting aspects of learning.
One important source of this ambiguity is the confusion in the literature between an
'outcome' and a 'process' perspective on organizational learning (Dodgson 1993, March
1994). Most definitions of learning tend to focus on outcomes of learning, as opposed to
what learning is and how these outcomes are achieved. Most contributions use the concept
when referring to a flexible organization, to proactive behavior, to organizational
intelligence etc. Because of dominance of the 'outcome' perspective, it remains rather
obscure how these improvements come about. Clearly, learning is a process and should be
studied as such. Just as research on individual learning should not be restricted to studying
intelligent people, neither should organizational learning be studied by only analyzing
efficient organizations.
1
During the 1990's, 184 articles on organizational learning were published in international journals.
This can be compared to 50 articles during 1980's, 19 during the 1970's, and 3 during the 1960's (Crossan
and Guatto 1996)
This observation has important implications for the study of organizational
learning. J ^ J h i s J h e ^ ^
is^defined as„J h ^ p r o c e s ^
L
knowfolge. Whether this process leads to improvement will be treated as a question to be
addressed rather than a presupposition.
Given that organizational life is often far from being rational (March 1995), the
same can be said for its learning behavior (March and Olsen 1976). Cognitive blinkers,
structural complexities, myopic forces etc. may all complicate the learning process.
Consequently, the outcomes of learning are often far from yielding an increase in
intelligence, creativity or whatever improvement the organization is seeking. Studying
organizational learning means addressing the processes along which organizational
knowledge is (re)constructed. The question of how to improve through learning should be
addressed by paying attention to the many traps and obstacles that can be found on the
road to 'successful' outcomes of learning. Awareness as well as the effort to overcome
these hindrances increases the possibility of meeting 'successful' outcomes.
One of the dangers of the present vagueness that characterizes organizational
learning is that it may soon be played down as just another buzzword of the fin du siecle.
To avoid such a premature death, theoretical clarity is needed. Consequently, the purpose
of this thesis is to help understand the concept of organizational learning. To be sure, no
claim is made here to present an alternative perspective on learning as being the only one
which is authentic, legitimate and valuable. Rather, the perspective on organizational
learning presented in this thesis should be read as one possible approach to the concept
which may stimulate and encourage the ongoing debate on organizational learning.
The general research issues behind this thesis comprise the following questions:
How to cope with the ambiquity that caracterizes the literature on organizational
learning, and what are the implications for organizations that want to improve
their learning?
The way I will approach this task differs at least in three respects from existing
literature:
2
1
The thesis borrows many of the fruitful ideas that already exist in literature on
organizational learning.
2.
Existing literature will be complemented with aspects that have been previously
overlooked or that have been given disproportionally less attention.
3.
The thesis is based on the idea that the process of organizational learning is full of
impediments which should be acknowledged whenever organizations are striving
for improvement.
This, then, in short reflects the structure of this work. After first having critically
reviewed the existing literature on organizational learning, an alternative perspective on
the concept is provided followed by its practical implications.
Every text has its personal touch and this present text constitutes no exception. The
study approaches organizational learning as a social phenomenon in which social
interactions between human beings are emphasized. Organizations are considered from a
perspective that in some respects is rather similar to the interpretive perspective, and
specifically to the constructivists' view on organization and communication (e.g. Putnam
and Pacanowsky 1982, Strauss 1978, Weick 1979). These authors assume that
organizational reality is socially constructed by means of communication. According to
these constructivists, an organization emerges through the interaction of people.
However, the ideas put forward here differ from a pure constructivist outlook in
that structural and institutional conditions are also emphasized which may interfere with
this construction of reality. Individuals are certainly not 'free' in (re)constructing whatever
knowledge they wish.
Secondly, the thesis goes beyond mere theoretical descriptions of processes that do
or do not construct organizational knowledge. In line with the functionalistic approach, I
will also elaborate on the implications for organizational practitioners as well as for the
role of information systems during learning.
Hence, the ideas that lie behind the thesis can be considered as a combination of
interpretive and functionalistic perspectives on organizations and on organizational
learning in particular.
3
A rather broad definition of organizations will be used. Organizations will be seen
as goal-oriented entities consisting of human beings interacting with each other, with their
identity as a product of these interactions. Multi-national corporations, departments, teams
and informal networks, etc., all fall within this broad definition.
Where the boundaries of these entities can be drawn depends on ones level of
analysis. Boundaries are subjectively perceived. Consequently, whenever I talk about the
environment, I refer to those people and organizations that do not belong to the
subjectively defined group of insiders.
Below a short overview of the basic arguments regarding the process of
organizational learning is given. Thereafter, I will clarify the research method that lies
behind this thesis. Finally, a short guided tour of the research is given.
1.2
FOUNDATIONS OF ORGANIZATIONAL LEARNING
The theory of organizational learning proposed in this work differs in many
regards from the traditional treatment of the topic.
First of all, in this thesis learning is approached from a higher level of generality.
The level of generality is considered by addressing the following question: can the concept
be based on single or multiple world views on organizations? Generic concepts can be
analyzed with multiple perspectives while perspective-specific concepts are more
connected to a single perspective. Examples of generic concepts are organizational
phenomena such as organizational culture, decision making, or communication. Examples
of perspective-specific concepts are, for example, management tools such as Total Quality
Management but they also include organizational phenomena such as planning and
2
control . In contrast to these perspective-specific concepts, generic concepts extend
beyond a particular type or form of organization, or beyond particular images,
perspectives, and metaphors. Decision making for example, can be analyzed from various
images. The same is true for organizational learning (Huysman et al 1993). A bureaucratic
2
For example, whereas the concept of Total Quality Management is based on an 'organic' image of
organizations, planning and control predominantly belong to the machine image (Morgan 1986).
4
view of organizations yields a completely different image of learning from a cultural
3
view .
Usually, organizational learning is approached as a perspective-specific concept. In
fact, organizational learning is most often seen as a management tool. With the use of
organizational learning, management is believed to be better able to turn the organization
into a more adaptive firm.
A perception of organizational learning as a generic organizational process differs
from many popular writings on organizational learning and 'the learning organization'
which use the concept as an expression of flexibility, change, and innovation (e.g.
Burgelman 1990, De Geus 1988, Garvin 1993, Senge 1992, Stalk et al 1992, Stata 1989).
These characteristics are often equated with organizational forms - such as the postindustrial organization (Huber 1984), the post-modern organization (Clegg 1990), and the
learning organization (Senge 1992) - that are able to counter-balance the mechanistic and |
bureaucratic principles that dominate many - industrial and modern - organizations.
At their most extreme position, the latest contributions to organizational learning
could be seen as a contemporary revival of Ford's principle of "the one best way of
organizing". Just as Scientific Management claimed that there was only one organizational
principle that would assure survival, popular writers on organizational learning assert
virtually the same position. It seems as if they have ignored the evolution of organization
theory that followed Scientific Management. Contingency theory - for example - taught us
that various factors such as task complexity, the environment, and technology influence
the way to organize. Later on, this theory was criticized for its deterministic view (e.g.
Pennings 1992, Scott 1990). Structures for example are viewed as static systems instead of
as a process (Giddens 1976, 1979, 1984, Strauss 1978), whereas technologies and
environments are viewed as determining organizational behavior without recognizing ¡
either social constructivism (Pinch and Bijker 1984) or enactment processes (Weick 1979). I
Hence, instead of one dominant organizational form that is tailor-made, a
I
perspective of organizational learning is needed that can tolerate many viewpoints.
3
The occurrence of general phenomena does not necessarily imply that when the social scientist tries
to analyze the phenomenon, this analysis is free of any subjective thought. On the contrary, social scientists
are human beings and as such have their own assumptions about "reality" (Kuhn 1970, Latour 1987). How I
have coped with this subjectivity in research method will be discussed in a separate section.
5
In this thesis, organizational learning is approached as a generic organizational
phenomenon that subsumes organizational learning techniques such as Business Process
Redesign, but also refers to processes such as for example imitating, experimenting,
simulating, networking etc.
The long history of organizational learning also justifies treating the concept from a
higher level of generality (Mirvis 1996). Klumper (1996) for example studied the
historical traces of organizational learning and concluded that the concept is perhaps just
4
1 as old as organization theory itself . In terms of formal organizations, this means that
organizational learning originated somewhere during the industrial revolution. In terms of
organizations in general, tracing its history hardly seems possible. Searching for its origin
is further complicated by the fact that organizational learning can be seen as a new term
5
coined to describe an ever existing process .
Furthermore, ideas about 'reasons' to learn - as addressed in this thesis - make the
concept of organizational learning more general than existing perspectives on learning do.
Traditionally, organizational learning has been articulated in the language of systems
theory. For example, Morgan (1986) refers to the metaphor of a brain as an information
processing system in order to describe organizational learning. Argyris and Schon (1978)
explicitly refer to various forms of feedback loops. March and Olsen (1976) talk about
environmental response and individual adaptation, whereas Senge (1992) bases his
learning ideas on Forester's theory of systems dynamics.
Although many learning processes are indeed activated by feedback information
with the aim of adapting to environmental changes, it is conceivable that learning can be
triggered by other things such as plain chance events, by experimenting, by the drive of
some active agents to 'actualize' themselves. Systems theory, although the dominant
theory where learning is concerned, is limited to a conception of learning as adaptation
4
One of the reasons why the concept is so fashionable may be that organizations experience a growing
need to become more adaptable and responsive to change. Secondly and related, as a result of the rapid tech­
nological changes, organizations face the need to learn to do things in a new and often radically different
way (Dodgson 1993a). Another possible reason is the growth of the "knowledge society" in which
knowledge becomes the key organizational resource (Drucker 1988).
5
Cyert and March (1963) were probably the first who referred to phenomena currently known as
organizational learning.
6
and as a result does not explicitly acknowledge these alternative triggers to learn.
In addition, the concept has often been approached as a two-sided phenomenon
(e.g. Argyris and Schon 1978, Fiol and Lyles 1985, March 1991, Miles and Randolf
1980, Miller and Friesen 1980, Senge 1990). Most of these contributions use this twosided aspect of organizational learning as a dichotomy. Either the organization engages in
single loop learning or in double loop learning (Argyris and Schon 1978), in lower level
learning or in higher level learning (Fyol and Lyles 1985), in adaptive learning or in
generative learning (Senge 1990).
By contrast, I will identify various ways in which an organization learns ranging
along a continuum from learning of well-known and well-tested ideas to learning new
knowledge or innovative ideas. An organization can learn from the knowledge dispersed
within the organization; it may learn from feedback information derived from the
environment; it may learn from the experience of other organizations; and organizations
may learn through the creation of new knowledge. A focus on one of the many ways of
learning will produce inefficiencies in the long run. Too much 'internal learning' produces
conservatism whereas too much 'creative learning' generates chaos and the inability to
learn from experience. This notion that organizational learning can manifest itself ^
differently and that there is no 'one best way' of learning, also contributes to a more j
general understanding of organizational learning.
Next and closely related to the level of generality, is the idea of learning as being
an integral part of organizational evolution. Again, this idea differs from mainstream
theories of learning. Many contemporary thinkers perceive learning as a strategic
phenomenon (e.g. Burgelman 1990, De Geus 1988, Garvin 1993, Senge 1992, Stalk et al
1992, Stata 1989). The concept of 'the learning organization' explicitly refers to
purposeful learning in order to be ahead of one's competitors.
I will argue however that most learning which appears at the strategic level is the
result of learning at the operational level. Ex post, it is conceivable to consider traces of
learning from experience as strategic avenues, although in practice, the boundaries
between operational and strategic organizational learning are blurred. A similar idea has
been put forth for example by Mintzberg (1988). According to this author, many so called
strategies are often not deliberately planned for, although in retrospect they are perceived
as such.
7
Given that organizational strategies often evolve out of traces from past learning
processes, organizational learning can be seen as a process of organizational evolution. An
evolutionary process refers to a history dependent, incremental process in which the
organizational past can be seen as imposing itself on the present through retention of
organizational experience in organizational knowledge. (Levitt and March 1988). This
history is not a straightforward process. Organizations face various problems and
unforeseen events during learning. As a result, the evolutionary nature of learning is what
may be charaterized as meandering (March 1990). Because the course of the evolution
depends upon the sequence of particular branches that are realized along the way,
organizational learning processes are not easily predictable. This has obvious implications
for strategic planning processes, and for the effort for radical change, "double loop
learning" (Argyris and Schon 1978), or "generative learning" (Senge 1990).
Its evolutionary character notwithstanding, organizational learning can also lead to
periods of revolution. In fact, this is what Argyris and Schon (1978) have called single
loop versus double loop learning. Whereas single loop learning refers to learning by
improving, double loop learning refers to learning by transforming. The evolutionary
concept of organizational learning proposed in this thesis has a more modest approach to
the idea of double loop learning. I do not want to deny that periods of revolution exist.
However, in line with the -incremental- innovation theories on organizational learning
(Nelson and Winter 1982, Rosenberg 1982, Sahal 1981), the process towards revolution is
one of evolution.
In addition to the questions why and how organizations learn, ideas on learning
that will be introduced in this thesis depart from most writings on learning where the issue
'who learns' is concerned.
The interest in organizational learning can be positioned along a continuum; along
this continuum range perspectives which treat organizational learning as an individual
phenomenon, to perspectives treating organizational learning as a collective phenomenon.
Whenever organizational learning is treated as an individual phenomenon, the only
learning that occurs within the organization is the learning of the individual members of
8
6
an organization .
At the other extreme of the continuum, organizational learning is something
different than the aggregate of individual learning. Learning can be regarded as a social
process rather than a psychological process. It was Emile Durkheim (1964) who first came
up with the idea that collective phenomena or 'social facts' as he labeled them, have to be
dealt with differently than individual phenomena. Social facts, like organizational learning,
arise out of human relationships and human association. Just as a fashion cannot be
reduced to individual cases without losing its essential meaning, so too organizational
learning cannot simply be reduced to individual learning.
In this thesis a middle of the road viewpoint is taken. The following observation of
Hedberg corresponds to the greatest degree with this viewpoint:
"Although organizational learning occurs through individuals, it would be a
mistake to conclude that organizational learning is nothing but the cumulative
result of their members' learning. Organizations do not have brains, but they have
cognitive systems and memories. As individuals develop their personalities,
personal habits, and beliefs over time, organizations develop world views and
ideologies. Members come and go, and leadership changes, but organizations'
memories preserve certain behaviors, mental maps, norms and values over time
(Hedberg 1981, p. 6)
To summarize the discussion so far, it can be seen that in this thesis organizational
learning is perceived as a rather fundamental organizational process. Several features can
be identified that are distinctive to this approach to organizational learning and that
characterize it as having a more generic character when compared to traditional theories of
organizational learning.
Learning is seen as a meandering evolutionary process that can neither be easily
planned or anticipated. In addition to this observation and closely related to it, reasons to
learn go beyond mere system theoretical explanations such as adaptation to environmental
changes. There are many triggers to learning that may be discerned, which also influence
the actual process of learning. There is not one best way of learning; every process of
6
For example, at a congress on "the learning organization", Utrecht 19 May 1992, the overwhelming
conclusion was that organizational learning is an Utopian concept. This was mainly suggested because
organizational learning was approached as essentially a collection of the learning of individuals.
9
learning may be valuable in its own right. Learning is considered as a mixture of various
processes of knowledge construction. Furthermore, organizational learning is perceived as
a dual process whereby the organizational code learns from the individuals in it just as
much as the individual learns from the organization.
1.3
RESEARCH METHODOLOGY
Given that the process of organizational learning has only sporadically been the
subject of theoretical examination, studying the topic calls for a methodology that allows
for theoretical exploration. However, whereas research methods which support the process
of merely empirical exploration have received increased attention, the same cannot be said
7
for methods supporting merely theoretical exploration . Nevertheless, general ideas which
govern conduct of empirical exploratory studies can also be seen to apply to theoretical
explorative studies.
The methodology of carrying out exploratory case studies and analyzing the
resultant research material has been addressed thoroughly (e.g. Agar 1986, Coenen 1988,
Glaser and Strauss 1976, Maso 1988). The general idea is that the researcher enters a
research field, for example an organization, with the aim of developing pertinent hypo­
8
theses and propositions for further inquiry . He or she starts without preconceived ideas
about what will be encountered or about how it will be interpreted and analyzed.
Ethnographic studies have proven to be a valuable research method where conduct
of exploratory case studies. Ethnographic research deals with the study of cultures, norms,
values, behaviors etc. and assumes that one may only come to understand these aspects
when they are observed by an outsider. Or as Kluckhorn once remarked: "It would hardly
be fish who discovered the existence of water" (in Wolcott 1975, p. 115). Thus, contrary
7
Exploratory research is seldom totally empirical or theoretical. During theoretical exploration, the
researcher makes use of insights obtained from existing theories as well as from first and second hand
observations from practice. What makes the two distinct is that empirical exploration is predominantly based
on ideas derived from empirical observations whereas theoretical exploration is merely based on ideas
derived from existing theories.
8
Given this ultimate aim of an exploratory study, this thesis will end with possible theoretical hypothesis
and recommendations for further research.
10
to the more generally accepted methods dealing with explanatory questions such as
surveys (Yin 1989), ethnographic studies cannot be programmed into a fixed research
design because the researcher has only limited initial knowledge about the research object
(Hammersley and Atkinson 1983). Of course, complete absence of prior knowledge is an
unrealistic condition. Indeed, some prior knowledge may even be useful when entering the
field of research.
"Good training in theory, and acquaintance with its latest results is not
identical with being burdened with 'preconceived ideas'. If a man sets out
on an expedition, determined to prove certain hypotheses, if he is incapable
of changing his views constantly and casting them off ungrudgingly under
the pressure of evidence, needless to say his work will be worthless. But the
more he is in the habit of moulding his theories according to facts, and of
seeing facts in their bearing upon theory, the better he is equipped for the
work. Preconceived ideas are pernicious in any scientific work, but
foreshadowed problems are the main endowment of a scientific thinker, and
these problems are first revealed to the observer by his theoretical studies"
(Malinowski 1922, cited by van Sluijs 1991)
What has been said about empirical exploration or ethnography in specific can be
translated into conditions of theoretical exploration. The lack of research standards that
guide the process of gathering and analyzing data is conspicuous for both types of
research methods. Furthermore, whereas empirical exploration starts with limited
knowledge of what will be studied because this would bias the research, theoretical
exploration starts with limited knowledge simply because this knowledge is lacking. But,
as Malinowski remarked, some knowledge of the domain of interest is needed in that it
helps to direct the search process and aids in interpreting the information one encounters.
Exploration requires an open-minded approach to the domain of research. During
empirical exploration, "to look a fool for the sake of science" (Kirk and Miller 1986, p.
49) yields the most fruitful insights. During theoretical exploration, this open-minded
approach is first of all supported when the research is characterized by pluralism.
Feyerabend (1975) is a well known advocate of such pluralistic methodology:
"A scientist who is interested in maximal empirical content, and who wants to
understand as many aspects of his theory as possible, will accordingly adopt a
11
pluralistic methodology, he will compare theories with other theories rather than
with 'experience', 'data', or 'facts', and he will try to improve rather than discard
the views that appear to lose in the competition. For the alternatives, which he
needs to keep the contest going, may be taken from the past as well. As a matter of
fact, they may be taken from wherever one is able to find them - from ancient
myths and modern prejudices; from the lucubrations of experts and from the
fantasies of cranks." (Feyerabend 1975, p. 47)
To illustrate the use of the research methodologies that lie behind the present
study, while conducting a merely theoretical, exploratory study, I will describe the
personal exploratory search process that marked the four years of conducting this
research
9
10
.
Research for this work started in April 1992 with the tentative title: "The use of
11
information and information systems to promote organizational learning" . Because I
had no idea what organizational learning really meant, I began the research with a
theoretical exploration of the concept. During the first half year I used the library of
various universities as my main source of information. Reading one article after the other
on organizational learning, I learned that the concept is extremely vague. Although a lot
11
has been written about the concept, a generally accepted notion seemed to be lacking .
This absence led to my first 'reseach breakpoint': the existing literature seemed to be
incapable of providing a clear enough notion of what organizational learning means, and
9
Such an explanation of the efforts that lie behind a study is most often subject of the first part of the
report: the preface. In this preface, researchers openly confess what a hassle the research has been, how
many obstacles had to be challenged, and how in practice the general excepted research-sequence was mixed
up. This practice seldom is conceived as important and is consequently treated as funny yet irrelevant
'preface-remarks'. This idea is expressed by Pettigrew (1985): "Contrary to the way the practice of research
is often taught and written up, the activity of research is clearly a social process and not merely a rationally
contrived act. Furthermore, it is a social process descriptively more easily characterized in the language of
muddling through, incrementalism, and political process than it is rational, goal-directed activity". As a
result, junior researchers learn from lies and learn to passed them on.
10
The reader will find a lot of reiteration when dealing with 'creative learning', discussed in chapter
seven of this thesis.
11
12
This work is totally financed by a research project with this tentatively approved title.
This search effort resulted in a paper on the various perspectives on the concept. See chapter two and
Huysman (1992).
12
without a notion of how organizations might learn, it is impossible to say something about
the use of information and information systems to promote it. Thus, I had to create my
own ideas concerning organizational learning.
If the existing literature does not bring you any further, other methods of research
are needed. Consequently, I decided first of all to tap another source of information:
personal communication with professionals. I talked with many friends and acquaintances,
and friends of friends and acquaintances about their ideas, knowledge, and experiences
with learning processes in their (client) organizations. My aim was not so much to learn
from them what organizational learning really meant but more to think through what could
be considered organizational learning and what could not. One such acquaintance inspired
me to tap still another source of information: his own behavior and the learning of the
organization for which he was working. This led to my first empirical exploratory study
13
on organizational learning . From this study, I learned many new ideas that had not
been reported yet in the research on organizational learning. Furthermore, I began to (
perceive organizational learning as something rather similar to innovation.
After this study ended, I had the opportunity to study the process by which a large
Dutch non-profit organization: "AZ", was becoming a profit organization. Since this
reformation had everything to do with innovating, I thought it would provide interesting
ideas on organizational learning. Consequently, I decided to spend more than half a year
analyzing the learning that took place within the company. After some months, however,
it dawned on me that no learning happened, at least not the learning that I had anticipated
or the learning that mirrored my initial definition. What to do? I could stop the project
and forget about my first impressions, or continue hoping something insightful would
eventually emerge. I decided to go for the latter option. Five months of research
frustration set in. Boxes full of recorded interviews and piles of written log-books and
impressions piled up, while I had no idea how to handle them.
Fortunately, 1993 ended with a pleasant interruption when the opportunity to visit
Stanford University for a period of nine months presented itself. I packed all my research
material and left for the United States, still hoping that finally something interesting would
materialize. My period as a visiting scholar began with attending the course "Advanced
13
Findings are described in 'Intermezzo IV of this thesis as well as in Huysman (1996b).
13
Organization Theory" given by Professor March. I thought this course would temporarily
free me of all the troubled thoughts on my research that had kept me busy for almost half
a year. The course and the articles discussed influenced me a great deal. The problems I
had been having with my research became transformed into 'foreshadowed' problems.
Every article discussed suddenly seemed to be relevant to organizational learning.
Subsequent courses had the same effect; there was learning within almost every aspect of
organizational behavior. Discussions with a large international group of researchers were
also extremely inspiring, as was the well-supplied university library. Although I still did
not know what organizational learning meant, I was sure that it went beyond notions like
j innovation.
Full of academic energy, I unpacked my boxes of research material and started to
look at them from scratch. During this analysis, I was advised by the indefatigable
Professor March. He taught me that there is a clear difference between an outcome
perspective (my prior conception of organizational learning) and a process perspective.
Furthermore, because of the irrational behavior of organizations, the outcome of learning
f
is often far from things like improvement, intelligence and radical change. This created
j
another 'research-breakpoint'. Suddenly, all my stray ideas on learning fell together. First
of all, approaching organizational learning as a process helped to clarify the concept since
it explains how organizations really learn, something that is missing in a lot of literature
on the concept. Secondly, it helped me to say something about the use of information and
information systems during learning. Finally, it made the problem of lack of innovation
that I encountered at AZ as irrelevant. In turn, various instances of learning processes and
especially problems with learning emerged out of my material.
Back in Holland, my inspiration continued. I finished the paper on the AZ
14
research and reexamined my first study by focussing more on the process instead of the
outcome of learning. It was time to write down all my ideas in the form of a thesis,
although this had to wait for some time after my (real) baby was born.
Writing is as much part of the exploratory research method as is the first period of
searching. Writing the thesis required a clear classification of all my ideas, something I
only had done for smaller aspects of the research such as during conferences. My head
14
14
The result of this analyzes are reported in "Intermezzo I" of this thesis, and in Huysman (1996a).
was full of ideas, derived from a mixture of personal communication, empirical studies,
personal experiences in organizations, literature on organizational learning, literature on
organizations in general, sociological literature, etc. etc. How to structure these thoughts?
Again, this painful experience was pleasantly interrupted by a planned wintersports
holiday. During the first two days I could not stop worrying about my thesis that was still
not written. I drew boxes, diagrams, arrows, etc on every paper I came across, even my
skipass had a draft of a possible table of contents. Fortunately, skiing took the upperhand
and on my way home, dozing off in the train, I suddenly saw the whole thesis in front of
me. Subsequently, I spent five months almost non-stop writing.
All these research efforts resulted in the present thesis which departs from the
original aim to explore the use of information and information systems in order to
promote organizational learning. The four years in which I had to finish the thesis were
not enough to provide the concept with theoretically sound arguments in order
subsequently to thouroughly analyze its requirements in terms of information systems.
Hence, because of time constraints, the present study is restricted to a theoretical
exploration of the concept of organizational learning. Although in chapter nine attention is
paid to the implications for the information systems discipline, this topic has not become
the central issue of the thesis. I do hope however that the thesis provides enough
theoretical understanding for future researchers to pursue this research project and study
the use of information systems to promote organizational learning.
Before leaving this section, it is necessary to describe what this research does not
provide.
The thesis lacks any (statistical) measures that mirror organizational learning
capacity. Such an effort would call for a variance approach in which all possible variables
that make up an organizational idiosyncratic nature are quantified and measured. Then, by
means of multi-variant analysis, these are set against a variable which measures
organizational learning capacity. There are several reasons why I have refrained from
carrying this out.
First of all, the circumstances that create an organizational identity are so diverse
that it is not possible to indicate them a postiori. Furthermore, measuring these conditions
15
in terms of quantifiable elements is almost a futile task. For example, how do we measure
the genesis of the organization without loosing its specific meaning? Thirdly, there is to
my knowledge no worthwhile standard that is able to measure organizational learning
capacity. What does it mean when organizations score high on its organizational learningcapacity? Does it mean that it is able to change quickly, or that it is able to assimilate new
knowledge in a short time? And what about the learning capacity in the long run?
Most importantly, variance analysis is not able to describe how this learning takes
place in practice. As mentioned earlier, an outcome perspective on organizational learning
perceives organizational learning as improvement in outcomes. The organization has
learned because it has improved its capacity for doing something. This outcome
perspective is however incapable of informing us how and why this learning takes place.
1.4
OVERVIEW OF CHAPTERS
In part one of this thesis the existing literature on organizational learning is
reviewed.
Although there is wide acceptance of the notion of organizational learning, no
theory or model is widely accepted. In order to provide some insight into this diversity,
chapter two highlights different perspectives and assumptions of the concept. Studies are
classified into six different perspectives: the adaptation perspective, the incremental
innovation perspective, the assumption sharing perspective, the organizational knowledge
perspective, the 'learning organization'
perspective and the social constructivist
perspective.
Chapter three gives a critical examination of the existing literature on
organizational learning by discussing its shortcomings. First of all, most writers link
learning with improvement while overlooking the fact that learning can also have negative
consequences and that frequent change can have its drawbacks. Furthermore, most of the
literature on organizational learning is too much focussed on the individual as an active
agent of organizational learning leaving the role of institutional aspects rather obscure.
Another shortcoming of the existing literature is that most, if not all, theories on learning
are based on some translation of systems theory. As a result, other triggers to learning
such as chance events are often neglected. In addition, theories are predominantly based
on a structural and rational model of organizations, neglecting the unpredictable and
16
irrational aspects of learning. Finally, the received theories on organizational learning
focus on only one or sometimes two types of learning, leaving alternative types of
learning untouched.
In order to balance all these biases, alternatives are given that underlie the
alternative perspective on organizational learning, presented in part two.
Part two of the thesis presents an alternative perspective of organizational learning
by way of introducing a typology of learning. This alternative perspective emphasizes the
process character of organizational learning. Four mutually dependent types of
organizational learning processes are introduced: internal learning, feedback learning,
learning from others and creative learning. Each of these learning types acknowledges the
inefficiencies of learning, the reciprocal relation between action and structure, the multiple
triggers to learn, and the possibility of unplanned learning.
In chapter four the process of internal learning is discussed. Internal learning can
be considered
as the backbone of all learning processes.
It deals with the
institutionalization processes in which individual knowledge is externalized and objectified
into organizational knowledge, while this organizational knowledge in turn is internalized
by organizational members. Various learning conditions will be discussed that should be
taken into account whenever organizations strive for positive outcomes of internal
learning. In order to illustrate the traps and obstacles of learning, a case story is presented
based on a qualitative case study. It deals with the mutual learning between two groups of
information
system designers.
The story
illustrates
the complicated
nature of
organizational learning and serves as a guideline scenario to which I will refer frequently
in this and in the next chapters.
In chapter five, the process of feedback learning is treated. Feedback learning
occurs whenever organizations learn from their own experience through environmental
reactions. The greater part of this chapter is devoted to processes that complicate feedback
learning.
Chapter six deals with the process of learning from others. Learning from others
occurs when organizations learn from the experiences of other organizations instead of
their own experiences. Various processes of imitation are described.
Chapter seven discusses the process of creative learning. This type of learning
deals with creating new knowledge. The chapter starts with the presentation of a second
17
case story concerning the inter-organizational networking of a group of people and in
specific of an information technology (IT) champion that led to the 'invention' of a
chipcard-idea. This story will be used to illustrate some of the theoretical arguments given
in this chapter. As is the case with the other three types of learning, creative learning will
not likely result in improvement, intelligence or innovation if it is not balanced with the
other types of learning as discussed previously.
Part three provides implications that can be derived from the theoretical arguments
put forward in the previous chapters.
Chapter eight has implications for organizational practitioners who seek to promote
successful learning. Successful learning refers to, as positively perceived outcomes of
learning processes, such as innovation, intelligence and improvement. First, it will be
argued that organizations should avoid the occurance of the various traps and obstacles to
learning as discussed in part two. These hinderances to learning are a result of focussed
information processing. Various causes of this so called "focussed learning" will be given
as well as some general ideas about how to avoid them. In addition to avoiding focussed
learning, organizations can improve their learning capacity by balancing and integrating
the four types of learning.
Chapter nine deals with implications for the information system discipline. Six
information intensive processes that characterize organizational learning will be reviewed
in terms of the role of information systems therein. Special attention is given to the issue
of how information systems contribute to problems of organizational learning and how this
could be avoided.
The thesis ends with concluding remarks, including a summary and implications
for further research.
18
PART ONE
IN SEARCH OF A THEORY OF
ORGANIZATIONAL LEARNING
In the next two chapters, existing literature on organizational learning will be
reviewed.
In chapter two, six perspectives are considered which can be identified within the
literature of organizational learning. I believe these perspectives are distinct enough to
treat them separately. Although each perspective has its own strength that should not be
ignored, they also have various weaknesses in common.
In chapter three, the existing literature will be critically reviewed by means of five
identified 'biases'. These biases concern hidden ideas and assumptions behind the
literature that have not been explained. These hidden assumptions or biases assure that the
received theories on organizational learning lean unnecessarilly in a certain direction
while overlooking others. I will try to indicate how these biases could be balanced.
From this search of the concept of organizational learning it follows that there is
no ready-made theory on organizational learning and, in fact, that an alternative more
integrated perspective on organizational learning is needed. Given that the six identified
perspectives all have their own valuable points, it would be a waste to ignore them.
Hence, a perspective is needed that integrates all these distinct ideas while at the same
time eliminating all identified weaknesses. In other words, through an alternative
perspective, one should be able to balance the biases distinguished in chapter three while
making use of the identified strong points of ideas already introduced, discussed in chapter
two. This alternative perspective is the subject of part two of this thesis.
CHAPTER TWO
DIFFERENT PERSPECTIVES OF ORGANIZATIONAL LEARNING
2.1
INTRODUCTION
In this chapter, most of the existing literature on organizational learning will be
surveyed. To provide insight into the common propositions concerning organizational
learning, studies are classified into six different perspectives on the subject: the adaptation
perspective, the incremental innovation perspective, the assumption sharing perspective,
the organizational knowledge perspective, the 'learning organization' perspective, and the
social constructivist perspective. These perspectives are believed to be distinct enough in
their approach to organizational learning as to treat them as distinct categories. To be
sure,, such an endeavor is subjective and may produce different outcomes when pursued by
15
others .
The purpose of this survey is to provide the reader with an insight into how the
concept of organizational learning has been developed so far. This knowledge enables the
researcher to position the alternative perspective of organizational learning, introduced in
part two.
Every section ends with a short critical note. Since a whole chapter will be devoted
on the weaknesses within the literature, I will only discuss the strengths of the various
perspectives.
2.2
ADAPTATION PERSPECTIVE
The notion of organizational learning as adaptation originated as an attempt to
answer questions raised by the contingency perspective. Viewed within this perspective,
organizations are treated as open systems engaged in exchanges with their environments.
The central proposition of the contingency theory is that organizational effectiveness is
directly related to the degree that internal organizational structures and processes "fit"
15
The identification of six perspectives is based on personal judgement obtained from an analysis of the
existing theories.
characteristics of the organization's environment. As environment changes, these
structures and processes must change to maintain this fit.
As a result of this insight, researchers have devoted considerable attention to the
question of how to design organizations that meet the demands of the environment. Much
of this work has focussed on how organizations deal with the complexity and uncertainty
presented by their environment (e.g. Woodward 1958, Burns and Stalker 1961, Lawrence
and Lorsch 1967, Galbraith 1973, Mintzberg 1979). It is generally accepted that require­
ments differ for organizations acting in a simple and/or static environment as opposed to
those acting in a complex and/or dynamic environment. Unlike the requirements of a
mechanistic organization acting in a certain and static environment, the requirements of
the organic organization dealing with an ongoing change of the environment raise many
important questions. How can an organization be consistently effective over time given
that changes occur in its environment? How is the fit between organizational structures
and processes and the characteristics of the environment obtained and more important,
maintained? With the rise of organizations acting in turbulent and uncertain environments,
these questions have gained prominence in the literature on organizational adaptation as a
learning process.
Some researchers have concluded that organizational learning occurs in response to
immediate problems, imbalances and difficulties much more than it does in response to
deliberate planning (e.g. Cangelosi and Dill 1965). By the identification of 'a performance
gap' as a major influence on learning (Downs 1966), organizational learning can be
considered as strategies to adapt to changes in the environment.
Cyert and March (1963) also perceive organizational learning as adaptation to
changes in the environment. This adaptation focusses on three different phases of the
decision-making process: adaptation of goals, adaptation in attention rules, and adaptation
in search rules. The behavioral theory of the firm assumes that organizations change their
goals on the basis of their experience. Goals are continuously adapted to incorporate the
experience of meeting previous year's goals, and also the experience of other
organizations in a similar situation. Adaptation in attention rules refers to the selective
attention that the organization gives to different parts of the environment. Organizations
learn to attend to some parts of the environment and ignore others. Similarly, adaptation
in search for solutions is also conditioned by previously tried solutions. Success reinforces
22
and failure discourages repetition.
March and Olsen (1976) provide an analysis of organizational learning under
ambiguity which incorporates limits to learning in organizations. They describe a model of
'simple complete cycle of organizational choice' in which the individual actions affect
organizational actions, which in turn affect environmental responses. The environmental
responses or acts affect the individual's beliefs and thus his/her behavior. This model of
choice serves as a tool for analyzing learning and adaptation by individuals and
organizations.
In the years following, March and his colleagues continued publishing on the
notion of organizational learning thereby extending the issue of adaptive learning in which
the emphasis moved from simple trial and error learning to organizational experience
captured in routines (Herriot et al 1988, Levinthal and March 1994, Levitt and March
1988, March 1988, 1991, 1995). Organizations learn by encoding inferences from history
into routines that guide behavior. The term routine is a broad one, including forms, rules,
procedures, conventions, strategies, and technologies around which organizations are
constructed and through which they operate as well as the structure of beliefs, frameworks, paradigms, codes, culture, etc.
March and his colleagues also address the notion of the ecology of learning:
learning is embedded in a mosaic of learning. Various learning units learn simultaneously,
and their learning also interacts. This makes organizational learning an extremely complex
and dynamic process.
Characteristic of all these contributions is that they address the complicated,
problematic and less-efficient side of learning. This is also one of the major strengths of
the perspective. Unlike most other perspectives that will be discussed, contributions of the
adaptation perspective are not averse to showing the inefficiencies of learning. The
learning behavior of organizations more often stimulate organizational inertia rather than
change, improvement and wisdom. Organizational learning is often path dependent. What
has been learned in the past is likely to direct future adaptation processes.
Focusing on the actual learning behavior of organizations is assured by the use of a
process perspective on organizational learning. This is in contrast to most other
contributions to the literature. Whereas organizational learning is mostly perceived from
an outcome perspective, equating learning with improvement and progression, the
23
adaptation perspective leaves the outcome of learning as an issue of investigation.
2.3
INCREMENTAL INNOVATION
The incremental innovation perspective deals with the process of innovation as
being one of incremental learning. Many writers have used the concept of organizational
learning to refer to the process of knowledge accumulation during innovation (e.g. Cohen
and Levinthal 1990, Nelson and Winter 1982, Pennings and Harianto 1992, Rosenberg
and Frischtak 1985 Sahal 1981). These writers all share the view that innovation embodies
the generation of new knowledge whereas prior knowledge is presumed to be an important
factor in determining whether an innovation is adopted and in what form. For the
diffusion of innovations to be successful, it is therefore crucial to determine what infor­
mation exists already in the organization.
Innovation is treated as an effort to bridge the distance between knowledge
currently available and the knowledge that an organization is striving to possess (Pennings
and Harianto 1992). By virtue of their unrelatedness with the current organizational
knowledge, discontinuous (McKee 1992) or big bang (Gluck 1985) innovations are more
likely not to be adopted, or if they are, likely to fail. This general assumption has induced
a large majority of innovation researchers to espouse an incremental notion of innovations
(Pennings and Harianto 1992).
Nelson and Winter (1982) are one of the best known proponents of an incremental
view of innovations. They emphasize that routines play a large role in innovations.
Routines spell out the appropriate activities and search for new knowledge. Consequently,
innovations are not discontinuous, but rather novel combinations of old routines.
As a result of these routines, each firm is relatively unique in accumulating
experience in the use of technology. This knowledge is mostly tacit, and is acquired in
problem-solving and trouble-shooting activities within the firm, remaining there in a
substantially uncodified state. In the words of Rosenberg and Frischtak:
"(E)ach individual firm is a focus where the progressive accumulation of
technical knowledge takes place, with production processes tending to
display many specific and idiosyncratic components" (Rosenberg and
Frischtak 1985).
24
Organizations are understood as entities which acquire the capabilities they have
through a time-consuming and incremental process of learning. The organizational know­
ledge, technologies or routines cannot be transformed from one organization to another
without recognizing their own internal history. The diffusion of innovations is not as
straightforward as some diffusion theorists seem to claim.
Sahal (1981) is another writer who treats the process of innovation as a process of
learning. Based on previous studies on technological innovations, he demonstrates that the
innovation activity is inherently full of uncertainties and problems. Given an environment
of rapid change in either demand or supply conditions as well as the unintended
consequences of human action, an explicit demand for a technological innovation cannot
be the single determining factor in its development.
Using various case-examples, Sahal demonstrates how often innovations disappear,
only to be conceived again at some later point in time when the related necessary knowhow becomes available. In between this period, dead-lock or progress takes place.
"A new technology does not emerge like Minerva from Jove's forehead.
Typically, it is the outcome of countless improvements in the capabilities of
some earlier, less specialized device through the gradual acquisition of
practical know-how. Success in technical problem solving is never just a
matter of armchair theorizing" (Sahal 1981 p. 111).
Innovation thus is considered a manifestation of learning by doing or learning by
experimenting.
Mckee (1992) has tried to combine the economic perspective on learning with an
organizational perspective. With the use of Argyris and Schon's (1978) concepts, he
shows how single loop learning, double loop learning and deutero learning are involved in
respectively incremental innovation, discontinuous innovation and institutionalization of
innovation. Firms engaging in incremental innovation must be supported by single loop
learning skills such as increased communication and increase of depth of contact with the
environment. Likewise, discontinuous innovation must be supported by for example
boundary spanners and slack resources. Institutionalization of innovation is supported with
for example structural cross-team contacts and continuous innovation training.
25
Like the adaptation perspective, this perspective focusses on evolution. Both
perspectives argue that history matters during learning. More than the adaptation
perspective, the incremental innovation perspective emphasizes the idiosyncratic nature of
organizations. It shows that there is no 'one best way of learning'. Every organization has
its own way of learning whereas the knowledge that has been institutionalized in one
organization cannot be transferred indiscriminately to other organizations. In other words,
the perspective
counterbalances
the optimism with which popular writings on
organizational learning stress the urge to unlearn past knowledge and to engage in
discontinuous change.
2.4
ASSUMPTION SHARING PERSPECTIVE
The central notion underlying the assumption sharing perspective is that of the
organizational 'frame of reference', which is somewhat analogous to Kuhn's (1970)
concept of a paradigm. These sets of beliefs or ways of seeing or organizing the principle
governing perceptions, are to a large extent particular to a specific organization. That is,
an organization is characterized by a paradigm that is shared by organizational members.
These paradigms provide a common language which makes possible the sharing of
experience and insights among organizational members. Although differently labeled, in
most of the literature on organizational learning this idea of the existence of a shared
frame of reference has been addressed. Here, it is sufficient to refer to the work of
Argyris and Schon (1978) which can be considered as a hall-mark for this perspective.
Although Argyris and Schon talk about "a detection of a mismatch of outcomes to
expectation which disconfirm organizational theory-in-use" (Argyris and Schon 1978, p.
19), this detection does not necessarily have to be adaptive. They distinguish incremental
adaptive learning (which they label single loop learning) from learning which affects the
fundamental organizational theory-in-use (which they label double loop learning) and
deutero learning (which means learning how to learn). Single loop learning occurs when
error correction proceeds by changing organizational strategies within a constant frame­
work or norms of performance. Double loop learning involves restructuring of
organizational norm and restructuring of strategies and assumptions associated with those
norms. It involves fundamental changes in the organizational frame of reference or 'theories-in-use' prevalent within the organization.
26
In their own words, organizational learning is described in the following terms:
"Organizational learning occurs when individuals, acting from their images and
maps, detect a match or mismatch of outcome to expectation which confirms or
disconfirm organizational theory in use. In the case of disconfirmation, individuals
move from error detection to error correction. Error correction takes the form of
inquiry. The learning agents must discover the sources of error - that is, they must
attribute error to strategies and assumptions in existing theoriesAn-use. They must
invent new strategies, based on new assumptions, in order to correct error. They
must produce those strategies. And they must evaluate and generalize the results of
that new action. "Error correction" is shorthand for a complex learning cycle"
(Argyris and Schon 1978, p. 19)
To describe the features of theories-in-use, Argyris and Schon developed two
models or ideal types: model I theory-in-use and model II theory-in-use. Model I has to
do with theories in use which inhibit double-loop learning. The four governing variables
of model I are: (1) achieve the purpose as the actor defines it, (2) win, do not lose, (3)
suppress negative feelings, and (4) emphasize rationality. The primary action strategies are
to control unilaterally the relevant environment and tasks and to protect oneself and others
unilaterally. The consequences of model I strategies include defensive interpersonal and
group relationships, low freedom of choice and reduced production of valid information.
These are negative consequences for learning because there is little public testing of ideas.
The hypotheses that people generate tend to become self-sealing. What learning does
occur remains within the bounds of what is acceptable. These defensive loops have been
called defensive routines (Argyris 1990). A defensive routine is any action or policy
which prevents experiencing embarrassment or threat and simultaneously prevents
reducing the causes of the embarrassment or threat.
The governing variables of the alternative model II include (1) valid information,
(2) free and informed choice, and (3) internal commitment. In this model, surfacing of
conflicting views is encouraged in order to facilitate public testing. The consequences of
model II action strategies include minimally defensive interpersonal and group
relationships, high freedom of choice and high risk taking. The likelihood of double-loop
learning is enhanced, and effectiveness should increase over time.
With its emphasis on self-reflection, there are striking resemblances between this
27
perspective on organizational learning and psycho-therapy. It is therefore not strange to
note that the theory of Argyris and Schon has proven to be a valuable tool for
16
organizational development practices .
The strength of this perspective lies predominantly in the notion of double loop
learning in relation to single loop learning. The concept stimulates the evaluation of actual
learning behaviors in terms of its revolutionary or incremental character. It shows that
most of the learning within organizations is of a conservative nature. Organizational
members tend to be blind in the face of norms and values that guide their behavior. The
notion of double loop learning also stimulates to think of self-reflection as a necessity to
challenge these cognitive blinkers.
2.5
ORGANIZATIONAL KNOWLEDGE PERSPECTIVE
The organizational knowledge perspective explicitly deals with learning as the
accumulation of organizational knowledge. Just as the work of Argyris and Schon (1978)
can be considered as a hallmark for the assumption sharing perspective as described
above, the work of Duncan and Weiss (1979) can be seen as representative of the
organizational knowledge perspective. Their ideas stimulated followers to link learning
with (computerized) information systems.
Duncan and Weiss (1979) argue that organizational effectiveness is determined by
the quality of the knowledge base available to the organization for making the crucial
strategic choices. Organizational learning then is defined as:
"the process within the organization by which knowledge about action-outcome
relationships and the effects of the environment on these relationships is developed"
(Duncan and Weiss 1979, p. 84).
Organizational learning is considered as a continuing evolutionary process whereby
16
Although they do not explicitly refer to organizational learning, Mason and Mittrof (1981) also
promote the idea of organizational self-reflection. They make use of the idea as a way to improve the
success of strategic planning. According to these authors, the barriers to this success is often deeply
embedded in an organization's social cognitive processes. In order to get a thorough evaluation of the
various cognitive models, they have introduced a method for stakeholder analysis and assumptions surfacing.
These self-reflecting methods allows the reflecting of espoused strategies against realized strategies. In
effect, these methods allows the detection of internal cognitive barriers to strategic plan implementation.
28
extension and or refining of the knowledge base is the outcome. These increments reflect
the addition of new statements of action-outcome relationships which are added to or
supercede existing statements.
Occasionally, however, this process is disrupted by 'paradigm revolutions'. These
revolutions are caused by experience of performance gaps which cannot be resolved within
the paradigm. The revolutions are somewhat similar to the double loop learning process
cited by Argyris and Schon (1978).
According to Duncan and Weiss (1979), knowledge is only organizational when it
becomes exchanged and accepted by others. In order to perpetuate this process, parts of it
are institutionalized in the form of formal learning systems and informal organizational
practices. Learning systems are the mechanisms by which learning is perpetuated in the
organization.
Examples of these
learning
systems
include
strategic
planning
systems,
management information systems, and informal arrangements like informal information
and communication networks.
The concept of learning systems brings the concept of organizational learning close
to that of (computerized) information systems. Jelinek (1979) is probably one of the first
who studied organizational learning systems. She examined organizational learning
systems at Texas Instruments which were used to manage the stream of innovative
products manufactured by the firm. The O.S.T. (Objectives, Strategies and Tactics)
system is a management planning and control system consisting of a series of linkages
between long range goals and shorter-range activities and the funding necessary to imple­
ment them. The long-range goals look to the fixture for ten to fifteen years; these are
broken up into short run business objectives for each business-unit of Texas Instruments.
The strategies provide guidelines for the coming three to four years. Finally the tactical
action programs detail the day to day activity with their current funding status. The
O.S.T. system is presented as an organizational learning system by which individual
insights and knowledge were institutionalized into a systematic procedure for successfully
29
17
managing the innovation of new products .
Further research on organizational learning systems was conducted by Shrivastava
(1981). He documented several organizational learning systems that were encountered in
sample organizations. These learning systems included a variety of formal, informal,
cultural, and historical schemes for managing the process of knowledge sharing within the
organization. Some of the learning systems were systems in the sense of formal
management information and control systems, others were systematic ways of viewing
organizational problems and sharing them with other organizational members. An
important feature of these systems is that they attempt to objectify the subjective personal
knowledge of individual members into an organizational knowledge base.
The organizational knowledge perspective is the only perspective that explicitly
deals with organizational learning as an information intensive phenomenon. Perceiving
organizational learning as such creates the possibility of communicating with the
information system-discipline. Not surprisingly, the work of Duncan and Weiss is most
often used in those few cases that organizational learning has been subject of explicit
information theoretical concern (e.g. Stein and Zwass 1995, Wijnhoven 1995).
2.6
THE 'LEARNING ORGANIZATION'
The learning organization perspective is perhaps the most popular within the
management and business literature at the moment. Two related perspectives fall within
this category. First, there is the literature that interprets the notion of organizational
learning as the idea that the organization needs a brain that will be able to think for the
rest of the organization. This is what can be called 'top level learning'. The other related
literature sees organizations as organized in such a manner as to be ahead of their
17
It is striking to note that these systems failed soon after her book was published. According to
Mintzberg (1989) this shows that learning can not be institutionalized: "..Texas Instruments' own fancy
planning system was subsequently believed to discourage innovation. In fact, there never was any evidence
that the company's success stemmed from anything more than a capable leader who knew how to learn and
whose own energy and enthusiasm enabled him to attract good people and to invigorate them. Good people,
of course, make for good organizations. They also design good systems, at least systems that are good for
them. But remove the good people and the systems collapse. Innovation, it turned out, could not be
institutionalized." (p. 350)
30
competitors. This organizational form can be called 'strategic learning'.
Learning agents of the top level learning perspective are for example the directors
(Garrat 1987), a management team (Stata 1989), or the company's senior managers
occupied with strategic planning (De Geus 1988). Garrat's main point of interest is the
role of directors in organizations. It seems that, although they are 'at the top', directors do
not know what is really going on within the organization while they are expected to know
everything. According to Garrat, this striking reality hampers organizational learning
processes since for an organization to be effective, it needs a permanent brain. The role of
director therefore requires a change in thinking as a specialist ('either/or'-thinking) to
thinking as a generalist fboth..and'-thinking). Further, the director must be able to cope
with more uncertainty and ambiguity and has to allow synergy between specialisms to
operate at the core.
According to De Geus (1988), fundamental changes in organizations' strategies or
major innovations depend upon the ability of a company's senior managers to absorb what
is going on in the business environment and to act on that information with appropriate
business moves. He defines organizational learning as:
"..the process whereby management teams change their shared mental models of
their company, their markets and their competitors" (De Geus 1988, p. 70).
The emphasis here is on opening up communication and acceptance of the idea that
the whole is larger than the sum of the parts.
An innovative aspect is the focus on the roles, skills and tools for leadership in
learning organizations.
The 'strategic learning' perspective became popular at the end of the eighties and
beginning of the nineties. Senge (1990, 1992) is perhaps the best known and most cited
representative of this theory. The best-seller of Peters and Waterman entitled "In search of
excellence" (1982) can be seen as the driving force behind the popularity of this
18
approach . In this book, bureaucratic, inflexible and rigid organizational forms are
18
Although the book can be seen as a model for the 'strategic learning' perspective on organizational
learning, Peters and Waterman do not explicitly refer to the 'learning organization'.
31
heavily criticized in order to make room for more open, flexible, lean, action-driven,
entrepreneurial organizational forms. These organizational forms encourage pro-active
behavior. Adaptation to changing environments no longer fit the present turbulent world of
today. What seems to be needed according to these authors are organizations that are able
to be ahead of their competitors and other threatening influences in order to create their
own future.
These ideas have been picked up by other management writers. The organizational
form that encourages pro-active behavior has been labeled 'the learning organization'
while the actions within these organizational forms are focussed on creativity, generative
learning, inductive reasoning, etc. (e.g. Garvin 1993, Pedler et al 1991, Senge 1992).
Although interested in the role of leaders in a learning organization as well, Senge
(1990, 1992) has a more modest interpretation of the brain-like function of the
management at the top. He picked up old assumptions of the theory on organizational
learning by emphasizing the distinction between adaptive learning and generative learning.
Adaptive learning is about coping with the environment and can be seen as the 'adaptive'
perspective on learning described in section 2.2. Generative learning is about creating as
well as about adapting. It requires new ways of looking at the world. Generative learning
will be reached by means of creative tension. The idea of generative learning can be seen
as an other formulation of Argyris and Schön's (1978) idea of double loop learning.
The learning organization concept has several strong points which may partly
explain its present popularity. One of the major strengths is that the perspective has
provided a bridge between theoretical, academic writings on learning and the practice of
organizations as perceived by consultants, managers and human resource practitioners.
Another strong point which is lacking within most other perspectives is its focus on
'generative learning'. Organizations not only learn in an adaptive manner, organizations
also learn more proactively.
2.7
SOCIAL CONSTRUCTIVIST PERSPECTIVE
Characteristic of the social constructivist perspective is its emphasis on practice.
Organizational learning cannot be fully understood unless its actual practice is studied in
depth. This micro-perspective results in a limited scope of learning. Although most
32
representative authors explicitly refer to organizational learning, in practice they talk about
the learning of small work groups or "communities-of-practice" (e.g. Brown and Duguid
1991, Ciborra and Lanzara 1994, Lave and Wenger 1991, Pentland 1992).
These authors argue that most theories on learning are based on transfer models of
knowledge or information. A major problem of these models is their neglect of the social
constructivist character of learning. According to the constructivist perspective, what is
learned is profoundly connected to the conditions in which it is learned. Consequently,
knowledge should not be isolated from practice.
This view on knowledge has its roots in American pragmatism (Dewey 1928,
James 1950, Mead 1934). Pragmatism contrasts the traditional ways of defining
knowledge. In general, knowledge is approached from a cognitive standpoint or from a
'structural' standpoint (Pentland 1992). The two following approaches are representatives
of the classical mind-body distinction (Rorty 1979).
The cognitive or mind approach equates knowledge with abstract representations,
and is according to Pentland a natural outgrowth of the traditional information processing
model of organizations (Galbraith 1973). It directs attentions to things like perception,
sense making, and belief.
The structural or body approach equates knowledge with organizational structures,
such as routines (Nelson and Winter 1982). The structural view offers the insight that the
capacity to act often depends upon things that are tacit. It directs attention to things like
objects, structures and routines (Pentland 1992).
In contrast to these mind or body approaches to knowledge, the social
constructivist perspective on learning approaches knowledge as consisting primarily of
situated performance. According to Pentland, we should:
"stop treating knowledge as a static entity that resides somewhere, like in a book
or in a library, and start treating knowledge as an active, situated phenomenon.
(Pentland 1992, p. 545) "
In a similar way, Brown and Duguid (1991) assert that in order to understand
learning, it is necessary to focus on the formation and change of the communities in which
work takes place. These communities are often unofficial or 'non canonical' and not
recognized by the organization. Most significantly, they are emergent: their shape and
membership emerge during the course of work practice and learning. Whereas work
33
practices are often the canonical way in which these communities emerge, learning is most
often non canonical. Brown and Duguid argue:
"Attempts to introduce 'teams' and 'work groups' into the workplace to enhance
learning or work practice are often based on an assumption that without impetus
from above, an organization's members configure themselves as individuals. (..)
people work and learn collaboratively and vital interstitial communities are
continually being formed and reformed. The reorganization of the workplace into
canonical groups can wittingly or unwittingly disrupt these highly functional non
canonical - and therefore often invisible - communities. (Brown and Duguid 1991,
p. 49).
Pentland (1992) examined how collective performances are accomplished in
practice. In his case study, software support hotlines services are implemented to assist
customers with technical problems. Individual support specialists often lack the personal
resources necessary to respond to a given call. There are several 'organizing moves'
available to them in order to resolve customer problems. For example, the call can be
given away or they can ask for help by asking a 'quick question'. Efforts to resolve
customer problems are in turn constrained by what is socially appropriate and what is
physically possible. These practical considerations limit the moves that are available to
service workers. Although organizing moves is limited to individual and work-group
learning, according to Pentland organizational learning would occur when the moves
members make are changing.
Ciborra and Lanzara (1994) assert that one important element of this learning is the
possibility of reflecting on these practices in what they call a 'formative context'.
Formative contexts are settings within which daily work routines are 'formed' and receive
their meaning and scope - for example through organizing moves. Formative contexts are
both action and history- based and have a fluid nature.
The social constructivist perspective has at least two positive aspects that are not
considered by other authors on organizational learning. First, there is the alternative
image of knowledge construction. Whereas organizational learning is mostly conceived of
as learning by gathering or acquiring knowledge, this perspective shows that knowledge
does not necessarily reside somewhere. Consequently, the perspective forces us to look at
34
the actual work processes. It is during these day to day activities that learning takes place.
Secondly, the perspective introduces the group as a level of analysis. Other perspectives
either look at the level of the individual or at the level of the organization.
2.8
CONCLUDING REMARKS
In this chapter, various perspectives on organizational learning have been
reviewed. All six perspectives have their own valuable points, as summarized in table 2.1.
Arranging the literature into separate perspectives is ultimately an arbitrary endeavor.
Therefore, the effort may rightly be accused of pigeon-holing the various studies on
19
learning, while neglecting their variability and uniqueness .
Perspectives
Main strengths
Adaptation
Inefficiencies of learning
Incremental Innovation
History matters
Assumption sharing
Self reflection.
Organizational knowledge
Information processing perspective
Learning organization
Generative learning
Social constructivist
(Group) learning during actual work-practices
^
)^
Table 2.1 Main strengths of the six perspectives on organizational learning
Interest in organizational learning that marks the discussions within organization
19
For instance, the contributions discussed under the heading of the organizational knowledge
perspective could have been discussed from an alternative perspective. Duncan and Weiss (1979) focus
primarily on the learning process of the dominant coalition as well as on the information flows for building
organizational learning. Consequently, it is somewhat arbitrary to position their study within organizational
knowledge perspective as it equally suits the 'top level' learning perspective present in some of the literature
on 'the learning organization'. The same goes for Jelinek's (1979) study on organizational learning systems
since these systems are management tools. More important however, the information which comprise this
O.S.T. system is definitely not restricted to the top level.
35
and management studies of today is predominantly focussed on two of the six perspectives
on organizational learning: the assumption sharing perspective, and the learning
organization perspective.
The assumption sharing approach has been popular ever since the book of Argyris
and Schon on organizational learning was published in 1978. The idea of single loop
versus double loop has especially proven to be of value in characterizing the learning
process of organizations. The 'learning organization' approach became popular at the end
of the eighties, and beginning of the nineties. Its popularity is mainly among practitioners
and conspicuously less among academics.
Up until this point, the review is free from any profound critical commentary. My
purpose was to provide, as objectively as possible, an overview of the various
perspectives on organizational learning that are present within the literature. Such an
overview makes it possible to demonstrate what the various contributions are within the
literature. Furthermore it implicitly shows where the existing literature falls short and
what it falls short of. In order to explain why an alternative approach is needed, chapter
three provides a critical review of the received theories.
36
CHAPTER THREE
A CRITICAL REVIEW
3.1
INTRODUCTION
A plausible approach to use when dealing with the concept of organizational
learning would be to exploit one or even a combination of all of the existing contributions
that have been discussed in the previous chapter. After all, until the present the various
theories have not been scrutinized by thorough critical review. Such an approach however
overlooks the various shortcomings from which all the six perspectives suffer. In this
chapter these shortcomings, present within the existing literature, will be reviewed in the
form of five biases.
The following biases will be identified: a bias towards improvement, a bias
towards individual action, a bias towards system thinking, a bias towards planned and
strategic learning, and a bias towards one or two-sided learning processes. It will be
argued that each of these biases steers the attention unnecessarily in a certain direction.
The arguments are based on theoretical and empirical findings which will be treated in this
chapter as well as in the rest of the thesis. Table 3.1 provides by way of a matrix a short
summary of this chapter.
An identification of biases calls for an introduction of an alternative which is as
much as is possible free of any preferences. Surely, this task may be virtually impossible.
What I do wish to pursue, is the idea that the existing theory is not complete and
unnecessarily tends to be slanted in certain directions. The alternative theory that will be
introduced in part two of this thesis is based on more 'balanced' aspects of learning.
adaptation
innovation
improvement
assumption
organizational
learning
social
sharing
knowledge
organization
constructivist
X
X
X
X
X
X
bias
individual
X
20
X
action bias
system thinking
X
X
X
bias
planned and
strategic
X
X
X
X
X
X
learning bias
one/two sided
X
X
X
learning bias
Table 3.1 Biases within the literature set against the six perspectives of
organizational learning
3.2
IMPROVEMENT BIAS
There is a tendency within the literature to equate learning with improvement,
intelligence, wisdom etc., or what I will refer to as 'successful' learning. This is
especially true for the assumption sharing perspective, the organizational knowledge
perspective and the learning organization perspective.
Organizations are believed to have learned when their performances have
improved. However, learning does not necessarily result in positive outcomes. This
becomes clear when we perceive organizational learning as a process instead of as an
20
In general, the incremental innovation perspective is too much focussed on the meso and macro level
of learning ignoring the part played by active agency in the course of learning.
38
21
outcome .
When organizational learning is studied as a process, the outcome of learning
remains one of investigation. To quote March, organizational learning should not be seen
as
"following a path of greater and greater elaboration, beauty, civility or fit with the
environment. The essential element is not that development leads to higher and
higher states but that it inexorably leads somewhere" (March 1990, p. 40)
There are various reasons why learning does not always lead to improvement.
Most importantly, organizational behavior is often far from being efficient and effective.
Unexpected events, myopic forces, and the confusion of history for example may
complicate learning processes (Levinthal and March 1993, March, Sproull and Tamus
1991). In part two and three of this thesis, I will elaborate more thoroughly on these and
other aspects that may yield inefficiencies and ineffectiveness as outcomes of learning.
Of course, the outcome of learning also heavily depends on the content of what is
learned. Just as learning from watching soap operas will probably not produce perfect
wisdom, so too should we be careful in equating organizational learning with improve­
ment. For example, it is hard to acknowledge that the learning about the Mafia necessarily
22
improves society, but it certainly involves a development .
In part two of this thesis I will elaborate on the process rather than the content of
learning as a hindrance for progression. Conditions such as selective perception,
miscommunication, too much homogeneity, etc. may all hinder successful outcomes of
21
Another reason why most popular contributions link learning with improvement might be due to their
optimistic and humane stance. Contributions to 'the learning organizations' and the assumption sharing
perspective emphasize a coalition model of organizations. Shared vision is needed as glue to hold people
together and give people the feeling of belonging. Shared participation, team building and individual learning
are the building stones of the organizations and ascertain a feeling of individual and especially groupresponsibility. Aspects of an arena-model of organizations (Strauss 1978) are hard to find. The driving force
behind organizational learning is solidarity instead of coercion and luring. Permanence of the organization is
assured through processes of self-renewal. In terms of McGregor (1960), these perspectives are based on
theory Y: a positive image of the organizational member who wants to put a lot of effort in keeping the
organization a learning organization. The notion of a learning organization consisting of a theory X image of
organizational members in which the idea that individuals are perceived as egoistic, lazy, and not willing to
learn is hardly conceivable.
22
This example also shows that what is considered positive outcomes by one group, for instance by top
management, does not necessarily apply to another group, for instance a group of employees.
39
learning, whether in terms of efficiency and effectiveness.
A process perspective also tones down the idea of radical change as the improved
outcome of organizational learning. Many contributions to the theory of organizational
learning perceive radical change as superior to incremental change (e.g. Argyris and
Schon 1978, Senge 1990, Swieringa and Wierdsma 1990).
However, too much changes produces chaos and instability; the organization is not
able to stand still and reflect on its past. In other words, too frequent changes inhibit
learning (Lounamaa and March 1987). Next to theoretical considerations, the practice of
learning also shows that most changes happen incrementally (Genschell 1997, Nelson and
Winter
1982, Rosenberg
and Frischtak
1985, Sahal
1981). The
idiosyncratic
organizational nature, its identity, or the existing organizational knowledge strongly
influence what will be learned in the future. In this thesis, the occurrence of path
dependency as a result of the history dependent nature of learning, will be considered an
important aspect of learning.
In contrast to popular writings on organizational flexibility, strategic change, and
organizational transformation, organizations seem to have an inherent bias to be
conservative. Various explanations for this conservatism in terms of organizational
learning have been given. For example, Senge (1992) argues that this conservatism is due
to the inability of organizational members to think in wholes instead of pieces. According
to Argyris (1990), difficulties with double loop learning are predominantly the result of
the defensive tendency among organizational members to protect themselves from open
confrontation and critique. Conservatism as a result of learning can also be explained by a
self-referential use of information. What information will be searched for and how it will
be interpreted in order to learn from it is largely determined by the organizational identity
(Huysman et al 1995).
March and Olsen identify four obstacles to learning (1976). Complete learning is
based on the 'complete circle of choice' (March and Olsen 1976) This model assumes that
individuals adapt their beliefs to environmental response. The change in beliefs or frames
of references lead to a change in individual action and will yield in turn a change in
organizational action which corresponds to the response of the environment. In practice
however, the 'rational' circle is often broken at one or more points. Incomplete learning
frequently
40
leads to reinforcement
of existing routines and consequently to the
reinforcement of the general frame of reference, ideologies, or belief systems. Even when
the existing routines are inferior, organizations continue to improve their competencies
within these procedures or technologies. This competency trap increases the likelihood of
persistence in inferior or outdated procedures.
As will be asserted in part two and three of this thesis, viewing organizational
learning from a process perspective reveals these and other problems and hindrances and
enables us to identify learning conditions which may produce more successful outcomes.
3.3
INDIVIDUAL ACTION BIAS
Although organizational learning has been studied both from a micro as well as a
macro approach, many organization theorists treat organizational learning at the level of
individuals and groups. They stress the individual action part of learning, neglecting more
structural considerations such as routines or performance programs. In fact, only some
'innovation' theorists (e.g. Nelson and Winter 1982) and some of the contributors to the
adaptation perspective (e.g. Cyert and March
1963) emphasize these structural
considerations. These efforts in turn however, tend to neglect the action part of the story.
The failure to distinguish between individuals and organizations as levels of
analysis is not only present in the discussion concerning organizational learning. Within
the general organizational literature, a theoretical pluralism exists concerning the interplay
between what Dawe (1970) refers to as "the two sociologies": one views individual action
as the derivative of the social system (such as structuralism and functionalism) and the
other views the social system as the derivative of individual action (such as the
interpretative sociologies). The "system argument" starts analysis with the organization as
a whole and locates individual action according to its place and function within the
system. The "individual argument" on the other hand begins with the individual and
procedes to find the system only as the aggregated outcome of individual acts.
The structure/action debate can also be perceived as a deterministic/voluntaristic
debate. Seen from the voluntaristic orientation, individuals and their created institutions
are autonomous, proactive, self-directing agents; individuals are seen as the basic unit of
analysis and source of change in organizational life. The deterministic orientation focusses
not on individuals, but on the structural properties of the context within which action
41
unfolds, and individual behavior is seen as determined by and reacting to structural
constraints. These constraints provide organizational life with an overall stability and
control (Astley and Van de Ven 1983).
For a long time, organizational theory has been dominated by a structural23
functionalist paradigm that emphasizes the deterministic orientation. This paradigm is
used here to refer to a broad range of positivist schools. In emphasizing structures,
structural-functionalists treat social phenomena as social facts, that is as concrete,
materialistic entities. Social reality exists "out there", external to the individual, and it
takes form prior to any human activity. Hence, these social facts impose on and shape the
behavior of individuals. Likewise, viewed from a structuralist-functionalist perspective, in­
dividuals are products of their environment. Individuals respond rather passively to
external stimuli. For these theorists,the primary unit of analysis is the organizational
entity; its social, psychological, and economic characteristics become static properties
rather than social processes (Putnam and Pacanowsky 1983).
A major shortcoming of the structural-functionalist perspective when dealing with
organizational learning is its neglect of the active behavior of individual members, or
"agents" (Giddens 1984). The organization learns but the link between this learning and
the behavior of the individuals 'within' the organization remains obscure.
Under the influence of interpretive sociology (Schutz 1971), the interpretive
perspective as representative of the voluntaristic orientation, emerged as an important
alternative organizational paradigm. In this perspective social reality is portrayed as
symbolic processes - created through ongoing actions and through meanings attributed to
these actions. Whereas the structural-functionalist perspective - with the contingency
theory as one of the prominent representatives in organization theory - views organiza­
tional structure as a static system of normative and/or behavioral relations - the
interpretive perspective views structure as process. Proponents of this conception are
Weick's social psychology of organizing (1979), Silverman's (1970) action theory and
Goffman's symbolic interactionism (1983).
23
When using the words "structural-functionalist perspective" I refer to a combination of structuralism
and functionalism. Structuralism deals predominantly with macro-level phenomenon and neglect micro
phenomenon. Functionalism sees phenomena as parts with a function for a larger whole. Although the two
have separate characteristics, they are often used in combination.
42
A major shortcoming of the interpretive perspective is the notion that structure is
an emergent property of ongoing action. This idea suggests that action unfolds free of any
preconceptions, and it underestimates the degree to which institutional patterns impose
prior constraints on the action from which structures emerge (Weick 1990).
A serious problem arises when these different schools of thought focus on only one
side of the issue and use such different logic and vocabularies that they do not speak to
each other directly (Astley and Van de Ven 1983). This problem can be reduced by
perceiving them as dualities rather than mutually exclusive pairs. In this way,
organizational processes are not seen as voluntaristic bottom up processes alone, nor as
complete deterministic top down structural processes either.
In a similar way, organizational learning is perceived in this thesis as taking place
through the action of individuals when these actions are simultaneously constrained by
institutional forces.
Although organizational learning is influenced by the activities of active agents, it
is at the same time a top down process. Organizational history, assimilated in
organizational memory, structures the activities of these learning agents. Thus, individual
learning is not free from any preconception. Institutional patterns such as organizational
norms and values, but also environmental rules and beliefs impose prior constraints on the
actions of agents. As a result, the learning within organizations is often conservative.
Because of this dual character between on the one hand the voluntaristic actions of
individuals - 'active agents' - and on the other hand the deterministic force of existing
organizational structures - 'structural properties' - organizational learning can be seen as a
process of structuration (Giddens 1976, 1979, 1984) or institutionalization (Berger and
Luckman 1966, Schutz 1971).
Although the 'Structuration theory' of Giddens provides an interesting perspective
on this reciprocal nature of social phenomena, I prefer the use of the concept of
institutionalization as approached by Berger and Luckman in their "Social Construction of
Reality" (1966). Unlike Giddens, these authors focus explicitly on the (re)construction of
knowledge and how this (re)constructed knowledge influences and is influenced by
subsequent (re)constructions.
Furthermore, a shortcoming of the Structuration Theory is that it is predominantly
43
centered on the level of individual action as well as (societal) structure, overlooking
intermediate levels such as in the case of organizations, the level of the group. Groups
play an important role in facilitating as well as discouraging processes of structuration.
Most often, individual action can only make a difference when this action is supported by
a group. For example, as symbolic interactionists - among whom Berger and Luckman
(1966) can be considered - have demonstrated, reference groups are influential
intermediaries between individual action and structural properties. They filter individual
beliefs and action by defining what is appropriate and what is not. Besides cognitive
support, groups are also vital in providing (political) support to individuals. During the
process of innovation for example, innovation champions need the support of a group of
loyal followers in order to make their efforts successful (Kanter 1983).
Analyzing organizational learning as a reciprocal process creates awareness that the
process cannot be considered as an entirely voluntaristic process in which (purposeful)
individual (interaction will lead to a change in the organizational memory or knowledge.
Individual learning agents are also constrained by structural properties in their learning
behavior.
Furthermore,
these individual actions have unintended
and unknown
consequences. This reciprocal character of learning will be discussed more theoretically in
chapter four when the core process of learning: "Internal learning" is treated.
3.4
SYSTEMS THEORY BIAS
Most, if not all theories on learning are based on some translation of system
theory. The organization adapts to changes in the environment (Cyert and March 1963), to
responses to organizational action (March and Olsen 1976, Argyris and Schon 1978), or
organizations are part of broader system of organizations that are all connected to each
other; learning means not thinking in pieces but thinking in wholes (Senge 1992).
In general, organizational learning is perceived as a way to correct errors and to
adapt to environmental demands. Organizations need to learn in order to adapt
successfully to environmental changes. The greater the uncertainty in the environment, the
greater the need for learning. Feedback information and information from external
environments are the keys to successful learning. In short, organizational learning is
mostly approached as an externally driven phenomenon.
44
As a result of this predominance of system thinking, other aspects of learning are
neglected. The trigger to learning may also be internally driven, in which case the system
thinking paradigm becomes less relevant. Organizations also learn for example from
organizational participants. This 'internal learning' has nothing to do with a need to adapt
to changing environments, or other forms of system thinking. Furthermore, organizations
may learn as a result of the need for managers to make a difference, or the almost
inherent drive of innovation champions to seek adventure. In a review of 'some of the
literature on organizational learning', Dodgson (1993b) for example observes that:
"(organizations, and the forms of collective and individual learning within
them, importantly affect learning processes and outcomes. Indeed, the role
of human agency and individual goals such as the drive for selfactualization are almost completely ignored in many accounts on
organizational adaptability. Organizational learning is stimulated both by
environmental changes and internal factors in a complex and iterative
manner" (Dodgson 1993b, p. 387).
As the two case stories presented in part two of this thesis illustrate, learning may
also result from unanticipated events which, again, a systems theoretical perspective would
tend to ignore. In section 3.6 I will return to this aspect of chance during learning.
I propose a wider scope of learning processes which allows for a broader range of
organizational learning triggers. In part two of this thesis a typology of organizational
learning is presented consisting of four mutually dependent types of learning. Although
these types of learning heavily overlap, they are conceptually distinct enough to treat them
separately. These four types of learning are internal learning, feedback learning, learning
from others and creative learning.
Internal learning consists of learning from existing knowledge within the
organization. A contemporary example of a planned way for internal learning is
knowledge management. Feedback learning deals with learning from experience through
the reactions of the environment. By focussing on adaptation, feedback learning is one of
the most generally accepted ways of learning within the literature. Learning from others
concerns learning from the experience of other organizations. It deals with (mutual)
imitation and can have a very subtle nature. Creative learning involves the creation of
45
knowledge and can often include experimenting. This type of learning deals with an
internally triggered introduction of variety in organizational knowledge.
Of all these four types of learning, feedback learning corresponds the most to the
systems theoretical explanation of organizational learning. In short, during feedback
learning organizations learn from their experience as a result of feedback information
derived from the environment. Other forms of learning are less driven by environmental
demands. This is certainly true for internal learning and creative learning. Both involve an
internally driven form of learning which is in the case of creative learning, often triggered
by chance, serendipity, cross-fertilization, and the individual drive for self-actualization.
3.5
PLANNED AND STRATEGIC LEARNING BIAS
Except for some contributions within the adaptation and the incremental innovation
perspective, many researchers portray organizational learning as an activity than can be
planned for. For example, Argyris (1990) argues that, in order to radically change basic
assumptions, defensive routines can be brought to the surface when open communication
sessions are organized. These organizational development tools can be designed
beforehand and could be used in various situations. In line with Gregory Bateson (1973)
the author refers to so called 'deutero learning' (second order learning) when dealing with
the institutionalization of these learning processes. Institutionalized processes of learning
can be found for example in research and development departments and planning and
marketing departments (McKee 1992).
Planned learning has also been a subject within the organizational knowledge
perspective. In general, authors within this perspective assert that information systems can
be build to support this deutero or institutionalized learning. As described in the previous
chapter, Jelinek (1979) as well as Shrivastava (1983) have analyzed organizational
learning systems that capture the information that organizations need to learn from the
environment.
The 'Learning Organization' perspective focusses mostly explicitly on planned and
strategic learning. Writers who make use of this perspective assume that organizations can
anticipate future learning behavior. And if learning can be anticipated, learning can also
be used for strategic purposes, such as gaining a competitive advantage. Organizational
learning is seen as a strategic activity, by which the ability to learn is considered an
46
important, even unique source of lasting competitive advantage (e.g. Burgelman 1990, De
Geus 1988, Senge 1990, 1992, Stalk et al 1992, Stata 1989). Given the increasingly
demanding environments, radical change is more desirable since it fosters progression,
innovation and change. As such, organizations can cope with the ever changing
environment. For example, the concept of "generative" learning introduced by Senge
(1990) occurs within "an organization that is continually expanding its capacity to create
its future" (1990, p. 14).
Learning organizations are 'built' in order to promote learning. Various design
criteria have been introduced that could foster organizational learning. It is argued that
whenever managers take these 'guidelines' into account, organizations will turn into more
efficient learning centers. In other words, organizations can anticipate their future learning
behavior.
The downside of all these optimistic contributions is that they tend to overlook the
more accidental and path dependent nature of organizational learning.
In the coming chapters I will discuss the power of history which can cause
conservative learning behavior. As a result of such forces, individual members or,
specifically managers, are not able to fully engineer the future (March 1990). Next to this
rule-following learning behavior, organizations are often confronted with internal as well
as external unanticipated events. These events limit the possibility of -strategic- planning
and deutero learning. Indeed, as the two case stories discussed in part II reveal,
unsystematic and unintentional learning is one of the common ways in which organizations
construct their knowledge.
The stochastic nature of knowledge construction has become accepted within the
literature of strategy formation (Mintzberg 1988, Quinn 1989, Vissers 1994). Quinn
(1989) for instance recommends 'logical incrementalism' thereby recognizing the cognitive
and process limits that constrain formal strategic practices. Such a process of strategy
formation allows for a continuous evolving process. Mintzberg and Waters (1985) propose
a continuum ranging from 'pure deliberate' to 'pure emergent' strategies. Emergent
strategy acknowledges the possibility of unplanned action; patterns develop in the absence
of intentions. Thus action can take place without previous thinking, courses of action
called 'strategy' - in retrospect - may not be based on previous plans. It is just a small
47
step towards acknowledging this rather unplanned nature of knowledge construction in
cases of organizational learning.
Several contributions within the adaptation and incremental innovation perspective
on organizational learning do indeed refer to the possibility of unanticipated learning (e.g.
March 1990, March and Olsen 1976, Pennings 1992, Sahal 1981). March (1990) for
example talks about a 'meandering' evolution thereby referring to the unplanned
unanticipated traces that learning process can leave behind. Organizations face various
problems and unforeseen events when learning.
"There are irreversible branches, thus path-dependence and decisive minor
moments. The branch-points, involving things like mutations, mating, com­
munication contacts, and fortuitous opportunities often seem almost chance
like in their resolution, yet decisive in their effects on subsequent history.
Though the path of developments is explicable in terms of a comprehensible
process, the realized course of natural evolution is difficult to predict."
(March 1990, p. 44).
Because the course of the evolution depends upon the sequence of particular
branches that are realized along the way, organizational learning processes are not easily
predicted - with obvious implications for strategic planning processes (Huysman et al
1994).
3.6
ONE OR TWO-SIDED LEARNING BIAS
Although it has often been argued that learning can be studied from various angles
(e.g. Dodgson 1993b, Easterby-Smith 1996), learning is less frequently seen as a multisided phenomenon. There is literature that focusses only on one side of learning, for
example the learning within organizations or the learning during innovation. But besides
emphasizing one particular type of learning, it has become a very standard endeavor for
researchers to approach organizational learning as a two-sided phenomenon (see table
3.2).
48
single loop vs. double loop
Argyris and Schön (1978)
exploitation vs exploration
March (1991)
adaptive learning vs generative learning
Senge (1990)
momentum vs revolution
Miller and Friesen (1980)
lower level learning vs. higher level
Fiol and Lyles (1985)
learning
reactive vs proactive learning
Miles and Randolph (1980)
Table 3.2 Organizational learning as a two-sided phenomenon
Perhaps the most common dichotomy is that of single loop versus double loop
learning (Argyris and Schon 1978) originating from cybernetics. Single loop learning is
concerned with controlling existing systems. Double loop learning questions the norms;
deviations from the norm will lead to a possible change of the norm.
March (1991) also draws a distinction between two organizational learning
processes, though without referring to pure system thinking. He uses the dichotomy of
"exploiting old certainties" versus "exploring new possibilities". Exploitation produces
reliable knowledge; exploration produces variety in knowledge.
Senge (1990) refers to the dichotomy between adaptive versus generative learning.
Whereas the first deals with learning by conforming to changing environments, the latter
refers to learning that produces pro-active behavior.
Fiol and Lyles (1985) make a distinction between 'lower level learning' and the
first process 'higher level learning'. Lower level learning pertains to changing behavior
while higher level learning deals with changing cognitions.
A more integrative perspective of organizational learning presented in this thesis
introduces organizational learning as a many-sided phenomenon consisting of four
mutually inclusive and dependent types of learning. Such an approach departs from the
tradition of approaching learning as a two-sided phenomenon, in three related ways:
49
1)
it does not assign a value to the various types of learning;
2)
it approaches the various types of learning as continuous and as being positioned
along a continuum ranging from learning of things already known to learning of
things not yet known;
3)
it argues that processes of organizational learning should incorporate elements of
various types of learning. As a result, promoting organizational learning processes
within organizations requires a balancing of various types of learning.
Many contributions to organizational learning consider 'higher level learning' as
superior to 'lower level learning', 'single loop learning' to 'double loop learning' and
'adaptive learning' to 'generative learning'. The distinction offered in this thesis does not
assign a value to one or another form of learning.
All types of learning may have their own value depending on the purpose of
learning. Organizations engage in internal learning for example in order to become more
knowledgeable about their dispersed experiences. Organizations learn from feedback
information as well as learning from the reactions of the environment. Organizations learn
from others for example because they do not have the expertise at home. Finally, creative
learning has its own value in creating new knowledge.
All four types of learning are of comparable importance. This observation has
significant implications that will be discussed more thoroughly in chapter eight.
Instead of dividing the various forms of learning into a dichotomy, I propose the
use of a continuum, ranging from learning of things already known to learning of things
not yet known. Thus, the organizational learning-dimension as shown in figure 3.1.
represents the degree of novelty or originality of knowledge that is introduced in learning.
Along this continuum "internal learning", "feedback learning", "learning from
others", and "creative learning" are positioned. A continuum to position the various
learning types is considered as more appropriate than a discontinuity such as a two-sided
approach, as there is no clear boundary between the various types of learning.
Figure 3.2 illustrates how the various forms of learning are related to increasing
the depth versus breadth of organizational knowledge. It should be noted that this
50
Internal
Learning
Feedback
Learning
Learning
From
Others
Learning of
things already
known
Creative
Learning
Learning of
things not
yet known
Figure 3.1 An organizational learning continuum
representation is ideal-typically. For example, the outcome of organizational learning
processes heavily depends on the effectiveness of learning processes. As will be argued in
later chapters, organizations tend to learn in a rather conservative way. This is also why
the depth of organizational knowledge as an outcome of organizational learning is
represented as occurring more frequently than an increase in breadth of organizational
knowledge as an outcome.
Internal
learning
Feedback
learning
Learning
from others
Creative
learning
Figure 3.2 Typology of learning in relation to the depth vs. breadth of
organizational knowledge
51
Of all forms of learning, the outcome of internal learning is likely to be the most
conservative. In other words, internal learning involves increasing the depth of
knowledge. Occasionally, internal learning may lead to an increase in the breadth of
organizational knowledge. For example, during day to day work practices, innovative - as
opposed to improved - ways of doing can be learned which may become externalized and
objectified into organizational knowledge (Brown and Duguid 1991).
Feedback learning may involve the increase of both the depth and breadth of
knowledge but is likely to be more of the first than of the latter type. This is mainly
because feedback information is a reaction of organizational action and as such does not
depart considerably from existing organizational knowledge.
Learning from others also involves the increase of both depth and breadth of
knowledge but is likely to be more of the second type since the diffusion of interorganizational knowledge will bring about a variety of existing organizational knowledge.
Creative learning is a way of learning that is most focussed on increasing the breadth of
knowledge. Of course, creative learning does not always result in an introduction of
variety in organizational knowledge. The case story about creative learning that will be
presented in chapter seven illustrates for example that no change in organizational
knowledge occurred, although the organization explored an innovative idea for more than
four years.
A final reason why this integrated approach differs from the existing literature is
that the various types of learning are believed to be overlapping and mutually dependent.
Although the four types of learning should not be approached as discontinuous
processes, I will treat them in part two of this thesis as distinct categories. This is done to
provide conceptual clarity. In practice, a focus on only one single type of learning may
have important negative tendencies.
I will argue that every type of learning should incorporate other types of learning
in order to overcome path dependency. This notion resembles a well-known thought of
Chinese philosophy, and in specific the idea that something would become its opposite
when allowed to develop to its extremes. In order to promote its development without
such negative tendencies, it must include elements of its opposite (Fung 1952).
Relating this notion to the present discussion of organizational learning implies that
all four types, although not being each other's opposite, should incorporate elements of
52
other forms of learning as to avoid extremity. For example, organizations which learn
from other organizations should create room for feedback learning, internal learning and
creative learning.
This idea has important implications for organizational practitioners who want to
promote successful outcomes of learning. The idea of balancing learning processes will be
discussed thoroughly in chapter eight.
3.7
CONCLUDING REMARKS
In chapter two, existing literature on organizational learning was reviewed by
categorizing the contributions into five different perspectives. It was argued that
organizational learning is predominantly viewed from one - or sometimes a combination of
two - perspective(s). While reviewing these six viewpoints on organizational learning, it
was concluded that they all have their own valuable points.
An alternative perspective could have been proposed by simply connecting the six
perspectives into one integrated theory. However, in this present chapter, it is argued that
the existing literature has also various important weaknesses. Five biases have been
identified which are to a greater or lesser extent present within the six perspectives. They
involve a bias towards improvement as outcomes of learning, a bias towards the individual
as learning unit, a bias towards system thinking as framework to analyze learning, a bias
towards planned and strategic learning, and a bias towards focussing on only one or two
types of learning.
Table 3.3 integrates the conclusions derived from chapter two and three.
53
Adaptation
Main strengths
Main weaknesses
Inefficiencies of learning
system thinking,
one or two sided learning bias.
Innovation
History matters
"structure bias",
one or two sided learning bias.
Assumption sharing
Self reflection
improvement bias,
individual learning,
planned learning bias,
one or two sided learning bias.
Organizational
Information processing
improvement bias,
knowledge
perspective
individual learning,
system thinking,
planned learning bias,
one or two sided learning bias.
Learning
Generative learning
improvement bias,
individual learning,
organization
system thinking,
planned learning bias,
one or two sided learning bias.
Social constructivist
Learning at actual work-
individual learning,
practices
one or two sided learning bias.
Table 3.3 Strength and weaknesses of the perspectives on organizational learning
54
In the four subsequent chapters, I will try to challenge the five biases by:
1)
providing a process perspective of organizational learning through a focus on the
way learning as a process takes place. The outcome of learning heavily depends on
the process of knowledge construction. In particular, the occurrence of desired or
successful outcomes of learning depends on the way organizations cope with the
many identified hindrances to learning as well as on the effort to balance the
various forms of learning. An awareness of the possible occurrence of these
24
learning-problems may increase the chance of 'successful' outcomes of learning;
2)
embracing the idea of reciprocity between individual action and organizational
structure by acknowledging that organizations learn from individuals while this
learning is influenced by the fact that individuals also learn from organizations.
Briefly, the idea of externalization, objectivation, and internalization knowledge,
taken from Berger and Luckmann (1966) forms the standard type of all possible
types of learning and will be discussed in chapter four;
3)
thwarting the system-thinking bias by showing that organizations do not only learn
by reacting to knowledge as input in a feedback loop. Rather, there are various
learning triggers, such as the will to imitate, the drive of individual actors to
'actualize' themselves or the occurrence of chance events. Feedback learning is
only one possible form of learning;
4)
emphasizing the power of the organizational past and the occurrence of
unanticipated events which may thwart the planned and strategic learning-bias. In
practice, revolutionary changes reflected in concepts such as "higher level
learning" (Fiol and Lyles 1985) or "double loop learning" (Argyris and Schon
1978) are pretty rare, and are often an end result of many small changes.
Organizations cannot just throw away old experience and begin over and over
again. The process towards revolution is often one of evolution. In addition,
organizational learning is often accidental. Internal and external unanticipated
24
Successful learning refers to learning-processes that are not obstructed by the various identified
problems of learning.
55
events complicate the planning of learning.
focussing on four conceptually distinct types of learning instead of only one or two
as is usually the case in the literature on organizational learning. Learning is
considered here as consisting of four different processes that are mutually
dependent. Depending on the situation, one or more types of learning may be of
more relevance. For example, in case the organization tries to learn from its
organizational members, internal learning dominates. In case the organization
imitates other organizations, learning from others is of great relevance. As will be
argued in chapter eight, too much emphasis on one of the four types of learning
will produce however path dependency. Furthermore, integrating elements of other
types of learning may contribute to the effectiveness of a particular type of
learning.
PART TWO
A TYPOLOGY OF ORGANIZATIONAL LEARNING
In the coming chapters an alternative perspective on organizational learning is
introduced. The aim is to provide a theoretical foundation to the concept of organizational
learning. I believe this is necessary since the concept still lacks a clear theoretical basis.
This theoretical exercise will be used in the final part of the thesis to reveal more concrete
implications in practice.
This part of the thesis integrates the different perspectives on organizational
learning as discussed in chapter two and challenges the five biases that have been
identified in the chapter three.
Four types of organizational learning processes are described in separate chapters:
internal learning, feedback learning, learning from others, and creative learning. This
typology reflects the possible ways in which organizations learn. More than just an
identification of various learning processes, the four types of learning are mutually
dependent. Too much focus on one of the types of learning may produce negative
consequences. Consequently, no type of learning is superior to other types of learning.
Hence, although creative learning is last in the row of learning-forms, its importance is
comparable to that of the other types of learning. This assumption is a substantial aspect
of the alternative perspective that is proposed in this thesis.
Although this idea of integrating the four types of learning forms part of the
theoretical arguments, it can at the same time be considered implications for
organizational practitioners who seek to produce successful outcomes of learning. Hence, I
will elaborate on this issue of integration when discussing the possibilities of improving
the organizational learning capacity in chapter eight.
By presenting a typology of learning, I integrate the various perspectives on
organizational learning that exist within the literature, as discussed in chapter two (see
table 4.1). Internal learning corresponds to the social constructivist perspective in that
both emphasize the social construction of organizations and the internal dynamics of
learning. It also borrows ideas of the organizational knowledge perspective by focussing
on the process of organizational knowledge construction. Feedback learning corresponds
to the adaption perspective in its emphasis on adapting to environmental responses. It also
corresponds to the assumption sharing perspective for its reliance on system-based
thinking and in specific on the detection and correction of errors. Learning from others
corresponds with the innovation perspective. Both address the diffusion of external
knowledge. Creative learning resembles the 'learning organization' school in that both em­
phasize the importance of generating new knowledge.
Typology of learning
Related perspectives
Internal
Learning from its members and the
organizational
learning
members from the organization.
constructing knowledge,
knowledge:
social constructivist:
situated learning
Feedback
Learning from environmental
adaptation:
learning
reactions
adapting to the environment
assumption sharing:
detecting and correcting errors
Learning
Learning from experience of other
incremental innovation: external
from others
organizations
knowledge diffusion
Creative
Learning through experimenting
the learning organization: the
learning
creation of new knowledge
Table 4.1 The origins of the four types of learning
Every chapter starts with a theoretical introduction followed by a discussion of the
possible traps and obstacles that organizations may encounter while learning. It is argued
that 'successful' outcomes of learning may be reached when these traps and obstacles are
taken into account as well as when organizations engage in balancing the various types of
learning. In chapter eight, while discussing the implications for organizational
practitioners, this need for balancing learning is addressed.
CHAPTER FOUR
INTERNAL LEARNING
4.1
INTRODUCTION
In this first theoretical chapter of part two, processes that can be described as
'internal learning processes' are discussed. Internal learning is considered here as the
basic or elementary form of all organizational learning processes. It deals with the
processes during which the organization learns from its members as well as the processes
during which members learn from the organization. As such, during internal learning,
organizations act as closed systems. In the course of the discussion in subsequent chapters,
this basic model of learning will be further elaborated in order to satisfy more complex
learning phenomena.
The chapter is conceived as follows: I will first describe what this concept entails.
Thereafter, I will approach internal learning as a process of institutionalization. This
process will be unfolded by distinguishing between externalization of individual
knowledge, objectivation and internalization of organizational knowledge. Subsequently, I
will discuss some of the traps and obstacles that are distinctive for internal learning and
that may complicate internal learning and organizational learning in general.
4.2
THE CONCEPT OF ORGANIZATIONAL KNOWLEDGE
Organizational knowledge refers to knowledge which an individual uses when
acting as an organizational participant. Much has been published about the concept of
organizational knowledge, although there still seems to be confusion about its meaning.
i
First, organizational knowledge may be seen as residing in formal descriptions of\
the organization and its activities or in the retained records of organizational activity. This j
type of organizational knowledge consists of formal knowledge about the organization and
may be viewed as analogous to the contents of an organizational knowledge base.
Examples of such formal organizational knowledge are the formal record of organizational
activity held in minutes of meetings, company reports, organizational mission statements,
financial information used in management accounting systems, organizational charts, etc.
^
^^^/oá
^
Rather than knowledge about the organization, organizational knowledge can also
be considered knowledge of the organization. Morgan (1986) for example discusses this
viewpoint when dealing with the image of a brain. Together with Ramirez (1983), he talks
of organizations as holographic systems in which organizational knowledge may be
embedded in their every component. With the growing popularity of organizational
learning, this idea of an 'organizational memory' has become subject of increased interest
(Sandelands and Stablein 1987, Stein 1995, Stein and Zwass 1995, Walsh & Ungson
1991). The concept is somewhat similar to the sociological conception of a collective mind
which as a construct evolved from the work of Durkheim at the end of the nineteenth
century. However, whereas collective mind refers to shared understanding and shared
interpretation, organizational memory does not necessarily achieve the same end. Current
literature on the topic has a rather functional perspective on organizational memory (Stein
and Zwass 1995). The operationalization of the concept is restricted to organizational
memory that allows for acquisition, retention, maintenance, search and retrieval of
information, leaving less structured organizational knowledge untouched. Organizational
war stories> dress codes, informal rules and routines, etc. cannot easily be collected,
retained and retrieved. Not only is most of this organizational knowledge tacit (Polanyi
1958), they are often not free of subjective interpretation and political bias (Orr 1990).
The concept of "organizational routines" (Levitt and March 1988, Nelson and Winter
1982) provides a possible solution to this problem of a too formal image of knowledge. In
other words:
"The generic term "routines" includes the forms, rules, procedures,
conventions, strategies, and technologies around which organizations are
constructed and through which they operate. It also includes the structures
of beliefs, frameworks, paradigms, codes, cultures, and knowledge that
buttress, elaborate, and contradict the formal routines. Routines are
independent of the individual actor who execute them and are capable of
surviving considerable turnover in individual actors." (Levitt and March
1988, p. 320)
Given that routines may be considered as restricted to tradition, customs and habit,
I prefer the use of the general concept of "organizational knowledge" thereby referring to
both formal and less formalized aspects of knowledge.
60
4.3
THE PROCESS OF INSTITUTIONALIZING KNOWLEDGE^
Internal learning can be perceived in terms of the process of institutionalization.
The essence of organizational learning is the construction of organizational knowledge
such
as
organizational
norms,
procedures,
technologies,
gossip,
etc.
Through
communication, individual knowledge may become collective (organizational) knowledge
while this accumulated knowledge will in turn influence subsequent action.
Given that all forms of learning that will be described in this thesis mutually
depend on each other, this process of institutionalization, although particularly relevant to
internal learning, also applies to other forms of learning.
The term 'institutions' is used to describe social practices that are regularly and
continuously repeated, are sanctioned and maintained by social norms, and have a major
26
significance in the social structure . Institutionalization is the process whereby social
practices become sufficiently regular and continuous as to be described as institutions. The
27
concept is widely used in sociology, though often without precise specification .
28
Scott (1987) distinguishes different 'institutional schools' . Institutionalization can
be conceived of as 'a process of instilling value'. Selznick for example argues that
"institutionalization is to infuse with value beyond the technical requirement of the task at
hand" (Selznick, 1957, p 17) which may lead to an unplanned and unintended nature of
institutions.
25
Until so far I have used the words knowledge and information interchangeably. However, as many
writers have tried to point out, the two concepts are not the same. It is impossible here to review all uses of
the two concepts, though some words are needed.
In general, information is about facts and symbols and can be communicated or transferred without
the necessary mediation of individuals whereas knowledge is more about know-how and cannot be uncoupled
from human beings. Von Hippel perceives know how as "the accumulated practical skill or expertise that
allows one to do something smoothly and efficiently" (Von Hippel, 1988). The importance of this definition
lies in the word "accumulated". Know how must be learned while information can be obtained. This means
that knowledge as know how is of more significance than information during the process of (organizational)
learning.
26
This description of the concept 'institutions' is obtained from the Dictionary of Sociology, N.
Abercrombie e.a., Pinguin Books second edition, 1988
27
Different schools of sociology treat the concept of institutionalization in different ways. For example,
functionalists tend to see institutions as fulfilling the 'needs' of individuals or society (e.g. Durkheim 1978,
Parsons 1960) while phenomenologists may concentrate on the way in which people create or adapt
institutions rather than merely respond to them (Berger and Luckman 1966, Schutz 1971).
28
In fact, Scott (1987) distinguishes four schools: two dealing with the process of institutionalization and
two with institutions as systems. I restrict this discussion to the process aspect of the theory.
61
Institutionalization can also be conceived of as 'a process of creating reality'.
Social order is founded on a shared social reality, which is created by social interaction.
When dealing with the process in which individual knowledge becomes organiza­
tional knowledge, I refer to the latter formulation offered by Scott (1987). When dealing
with the process in which individuals make use of the organizational knowledge in order
to act as an organizational member, I refer to his first formulation.
Berger and Luckman refer to three phases or "moments" that can be distinguished
in the process of institutionalization: "externalization, objectivation, and internalization"
(Berger and Luckmann 1966). Externalization is the process in which personal knowledge
is communicated to others. Through externalization, "society becomes a human product".
Objectivation is the process during which "society becomes an objective reality".
Durkheim (1964) considers these objectified behavior patterns as "things".
During the moment of internalization, "the objectified social world is retrojected
into consciousness in the course of socialization" of the individual.
"Through
internalization man becomes a product of society" (Berger and Luckman 1966 p. 79).
As such, the authors refer to a dialectical relation between action and structure:
"The relationship between man, the producer, and the social world, his
product, is and remains a dialectical one. That is, man (not, of course, in
isolation but in his collectivities) and his social world interact with each
other. The product acts back upon the producer." (Berger and Luckman
1966, p. 78).
The "moments" of Berger and Luckman correspond to a certain extend to Giddens'
structuration theory (1976, 1979, 1984).
Giddens is one of the most well-known contemporary sociologists who proposes a
dialectical relationship between action and structure. Action and structure pre-suppose
each other, instead of being mutually exclusive. Giddens is more explicit than Berger and
Luckmann about the possible occurrence of consequences of human action that are
unknown or unintended.
Figure 4.1 depicts in a highly simplistic form this institutionalization process when
related to organizational learning.
62
Externalization
Organizational knowledge
Individual action
<
Internalization
/
Expression
Individual beliefs
Figure 4.1 Internal learning as a process of institutionalization
Individual action involves the behavior of individuals and may also include
expressed beliefs. Although I refer to the individual, these individual actions do not
exclude collective action such as group action.
Organizational knowledge refers to formal and informal organizational aspects such
as organizational paradigms, technologies, procedures, norms, values, strategies etc, as
described in chapter one. The use of this knowledge assures organizational action. Thus,
whenever speaking of organizational action, I refer to individuals making use of
organizational knowledge while acting or thinking.
Organizational knowledge in turn influences the individual beliefs through the
process of internalization. With individual beliefs I refer to individual theories regarding
the way of acting and thinking. Individual beliefs are the unexpressed attitudes and
29
opinions . Again, although referring to the individual, shared understandings and shared
29
There is much ambiguity around concepts such as individual attitudes, individual opinion, individual
ideology, etc. Attitudes are treated here as more or less well organized systems of ideas, which are affect
laden and may have direct implications for action.
63
30
beliefs are not excluded . It is crucial that these beliefs belong to the individual or group
of individuals and are not necessarily part of the organizational knowledge.
Together, individual beliefs and action form 'individual knowledge'.
Organizational knowledge is of course not the only source that influences this
individual knowledge. Personal experiences such as (previous) work experience,
education, cultural background etc. are all important influential forces which create unique
individuals and as such are important sources of variance.
Externalization occurs when individual knowledge is shared among individuals.
When this externalization results in organizational knowledge, I refer to objectivation. The
process of internalization occurs when individual actors integrate this organizational
knowledge into their personal beliefs. Finally, expression is the process through which
individual beliefs are put into practice. Given that this latter process mainly deals with
individual learning, I will restrict the following discussion to the first three processes in
relation to organizational learning.
4.3.1 Externalization of individual knowledge
Through communication, individual knowledge can be externalized. In the words
of Berger and Luckmann (1966), through externalization, the organization becomes a
human product.
With the growth of knowledge-workers and the increase of worker-mobility,
organizations and managers in particular increasingly feel the need to be continuously
informed about the knowledge that is present within the organization. This issue is of
special importance within 'knowledge intensive firms' and professional bureaucracies such
as consultancy firms and universities in which knowledge workers are predominantly
30
In fact, it can be argued that individual beliefs are always socially constructed. This idea of socially
constructed beliefs has been advanced by Mead (1934). He argues that individual beliefs are created by
engaging in 'internalized conversations between self and others'. Such conversations require taking the
perspective of (significant) others while not all interactants need to be "in separate bodies" (Weick, 1979, p
100). "It is in the form of the generalized other that the social process influence the behavior of the
individuals involved in it and carrying it on, i.e., that the community exercises control over the conduct of its
individual members; for it is in this form that the social process or community enters as a determining factor
into the individual's thinking.... And only through the taking by individuals of the attitude or attitudes of the
generalized others toward themselves is the existence of a universe of discourse, as that system of common or
social meanings which thinking presupposes at its context, rendered possible" (Mead 1934, p. 155).
64
professionals who feel more committed to their personal projects and clients than to the
organization as a whole (Schon 1982). Lately, this need to support the externalization of
private knowledge has been considered an important aspect of what has been called
"Knowledge management". Several consultancy firms for example introduced knowledge
management information systems that capture the knowledge within the organization. I
will return to the concept of knowledge management systems in chapter nine, when
dealing with their implications for information systems.
Externalization of individual knowledge can take place in a variety of ways,
depending on the combination of explicit and tacit knowledge, and the richness of the
communication medium used to externalize knowledge.
When talking about knowledge and organizational learning, it is important to keep
the distinction between tacit and explicit knowledge in mind. Polanyi (1958) addressed the
basic question: why do individuals know more than they can express? That knowledge can
be tacit has broad implications for understanding learning, and in particular for the
diffusion of individual know-how and individual beliefs. These can be so ingrained that
they are taken for granted (Nonaka 1988).
Knowledge expressible in language, is only the tip of an iceberg of our knowledge.
According to Polanyi (1958), there are two categories of human knowledge: articulable
knowledge that is transmittable with a formal, systematic language; and tacit knowledge
that is extremely personal, not formalized and difficult to communicate. Clearly, factual
knowledge as in information can be externalized much more easily than tacit
31
knowledge .
Next to the type of knowledge, Daft and Lengel (1986) propose that richness of the
communication media selected is closely linked to the learning in organizations. They
characterize media as high or low in richness based on the capacity to convey information,
31
The importance of tacit and explicit knowledge during organizational learning has been addressed by
Nonaka. His ideas on organizational learning are based on an innovation-perspective of 'knowledge
management' (Hedlund and Nonaka 1993, Nonaka 1988, Nonaka and Johansson 1985). His arguments build
on the premise that the generation and exploitation of knowledge in an organizational context revolves
around the interplay of explicit and tacit knowledge (Nonaka 1988).
Although his ideas are enlightening, they have a rather weak theoretical explanation. In this and
coming chapters, I will try to provide a more theoretical based explanation for the fundamental process of
organizational learning.
65
whereas information is defined as that which can change a person's understanding or
mental representation. Consequently, media richness is defined as the medium's capacity
to change understandings within a specific time interval.
Assumption sharing for example, takes place through exchange of opinions,
perceptions, and judgments. People can bring different frames of reference to the
discussion, so disagreements need to be brought to the surface and resolved. Rich media
such as face to face communication and group meetings, are better able to support the
construction of shared cognitions and to resolve equivocality through discussion (Daft and
Lengel 1986).
4.3.2 Objectivation of knowledge
Whenever this externalization is answered by a confirmation of dominant coalitions
32
within an organization or group, one could speak of objectified knowledge . Dominant
coalitions may be formed by senior management who can be seen as the gatekeepers of
formal organizational knowledge. Dominant coalitions may also be formed by a critical
mass of organizational members. Whereas in the first case the objectivation of knowledge
is primarily influenced by exercises of power, in the latter case objectivation may also be
influenced by other social psychological forces.
Objectified organizational knowledge is knowledge that is 'accumulated' in the
organizational memory. The organizational memory has been defined by Walsh and
Ungson (1991) as "stored information from an organization's history that can be brought
to bear on present decisions". In a similar vein, Stein and Zwass (1995) consider
organizational memory to be the means by which knowledge from the past is brought to
bear on present activities. Later authors make use of a rather formalized and structured
image of the concept of organizational memory. I prefer a looser concept, referring also
to less structured aspects such as stories, dress codes, etc.
The words objectivation and organizational memory may evoke images of
knowledge which is stored somewhere, for example in manuals, in technologies, in the
32
The adjective 'objectified' when referring to organizational knowledge is in fact redundant since
organizational knowledge is always brought to a level higher than individual knowledge.
66
heads of individual members etc. However, organizational knowledge can also be more
active, as it is created, adjusted and changed in action. Furthermore, the words
'organizational knowledge' are somewhat misleading since this 'situated knowledge' (Lave
and Wenger 1991) is mostly shared by a small collective such as a team of repair men
(Orr 1990), a computer call center (Pentland 1992) or a group of information systems
designers (Ciborra and Lanzara 1994).
Thus, whenever I speak of organizational learning, it depends on the unit of
analysis whether the group or the organization is referred to.
4.3.3 Internalization of knowledge
Through internalization, individuals become and stay organizational members.
Internalization has been described by many organizational theorists when dealing with
socialization processes or processes of enculturation (Schein 1992). Internalization
essentially means becoming an "insider".
Internalization of organizational knowledge can be supported with the use of
structured methods such as manuals, training courses, organizational reports, etc. Caution
is however needed when relying too much on such formal descriptions of organizational
practices or formal teaching modes. "It can lead to the isolation of learners, who will then
be unable to acquire the implicit practices required for work" (Brown and Duguid 1991,
p. 48).
An important means of internalization informal methods is the exchange of stories.
Stories serve an important role in internalizing knowledge that is 'noncanonical' instead of
'canonical' (Brown and Duguid 1991). Canonical practices refer to espoused practices
(Argyris and Schon 1978); they are formal descriptions of work, abstracted from actual
practices. Noncanonical practices refer to the actual practices taking place in
organizations. In other words, descriptions of canonical practices are based on the opus
operatum, the finished view, while noncanonical practices are based on the modus
operandi, the way a task, as it unfolds over time, looks to someone at work on it
(Bourdieu 1973).
Orr (1990) for example, has conducted ethnographic research on the noncanonical
practices of service technicians (reps). He concludes that these 'reps' frequently make use
of stories in order to fill the gap between the canonical descriptions of practices found in
67
manuals and other forms of "directive" documentation, and the actual (problematic)
situations that occur in practice. Through story telling, reps exchange their personal
experiences and are able to diagnose problematic situations - in this particular case a
troublesome machine.
For the reps, learning-in-working or learning by doing is an occupational necessity.
Their actual work practices are similar to Levi-Strauss's concept of 'bricolage': the ability
to "make do with 'whatever is to hand'" (1966, p. 17). A similar observation is given by
Hutchins (1991) in his analysis of navigation teams in the US Navy, and by Hirschhorn
(1984) in his analysis of computer operators at Three Mile Island. In both cases,
understanding of the task at hand is constructed within teams of operators through forms
of bricolage.
Lave and Wenger (1991) introduce the concept of "legitimate peripheral
participation" as a way to promote a noncanonical way of internalization. It deals with the
'rightful' possibility to participate in action, in order to learn the practice in detail. Brown
and Duguid (1991) argue for more awareness of this type of learning:
"It is a significant challenge for design to ensure that new collaborative
technologies, designed as they so often are around formal descriptions of
work, do not exclude this sort of implicit, extendable, informal periphery.
Learners need legitimate access to the periphery of communication - to
computer mail, to formal and informal meetings, to telephone conversations,
etc. and, of course to war stories. They pick up invaluable know how - not
just information but also manner and technique - from being on the
periphery of competent practitioners going about their business" (Brown
and Duguid 1991 p. 50).
Up until this point organizations have been portrayed as closed systems.
Individuals learn from organizational knowledge while at the same time the organization
learns from the individuals. The only variance - introduced at the level of the organization
- arises through the hiring of new members, through the gradual change of individual
beliefs, and the combination of individual knowledge skills.
In the following chapters, other types of learning will be discussed that explicitly
68
deal with learning processes which may in fact increase the breadth of organizational
knowledge. For now, I will continue this chapter on internal learning with an analysis of
the possible traps and obstacles that may complicate the process of internal learning.
However, before continuing this theoretical exertion, it is interesting to take a look
at a case story on organizational learning. The story is based on a qualitative case studies
33
that have been conducted at a Dutch company: "AZ" .
In fact, the story concerns the various processes of learning that took place at an
information system design department. It mainly illustrates the problems of learning the
department was facing or often did not face. It provides several anecdotes of learning
processes that produced inefficiencies. As such it illustrates that learning, when
approached from a process perspective, does not always result in positive outcomes such
as more intelligence and more efficiencies.
Ethnographic research methods were used based both on interviews as well as
observations. The study was conducted from July 1993 until December 1993. I observed
the group of system designers for three days in a week on average.
During six months of research, I had interviewed almost all of the people
employed at section A and half of the other department members to which section A
belonged. Most of these interviews were repeated again after several months. The
interviews had an unstructured character; I asked people to reflect on their experiences in
order to delve more deeply into the individual perceptions of the situation. All interviews
were tape recorded and fully transcribed. Aside from interviews with the personnel
manager, department managers, and information system designers, information was
obtained from archival study, observations of five plenary meetings, and participation in
social events such as drinks, lunches, "outings", etc. Important sources of information
were gossip and idle talk. Especially after a month of getting used to each other, people
started to perceive me as a confidante. It was predominantly out of these informal
conversations that ideas of learning emerged.
In this chapter as well as in the following three chapters, anecdotes of the story are
33
In order to maintain the anonymity of the people, the organization and the sections have been given
different names.
69
34
used to enliven the theory with practical illustrations .
INTERMEZZO I
LEARNING FROM NEWCOMERS
35
ATAZ
The hiring of newcomers can be an important trigger to learn. Newcomers look at
the organizational world afresh and may see imperfection, inadequacies, and weaknesses
that 'oldtimers' do not see anymore. Newcomers can also bring in new ideas which may
be a significant source of variation. Although newcomers may stimulate change,
conservative tendencies within the organizations often block this learning process.
AZ is one of the largest non-profit service providers in the Netherlands. The
company can be typified as a paternalistic bureaucracy although at present, because of
commercialization processes things are changing significantly. In the past, AZ provided
life-long employment for its employees; turnover was always a rare phenomenon. Possible
follow-up training courses were all taken care of by AZ and until recently, newly hired
employees were trained at the company school. In short, during its hunderd and fifty years
of existence up until recently, AZ provided security, certainty, and a future. For marry
employees, this perceived "soft-cushion" identity was an important reason to apply for a
job at AZ.
The information system design department employed seventy people and came into
existence through the division of a former computer department focussing primarily on
programming, into a programming and a design department.
Necessary criteria for the job of information systems designer were primarily based
on years of appointment at AZ. Although some in-house training courses in information
systems design were offered, most designers continued using the same standards that
guided their previous job as computer programmer. For example, programming was done
34
Although the story as well as the story described in Intermezzo II, chapter seven, are used in this
thesis only to enliven the theory, conducting the two case studies and analyzing the material has been an
important source from which the present theoretical arguments enfolded. Hence, the two case stories should
be considered as results of two exploratory case studies that have been substantially contributed the theory.
35
70
This story is a short version of the paper "Dynamics of mutual learning" (Huysman 1996a).
more or less in isolation; communication exchange among colleagues, the manager, and
the customers was limited. The learning that occurred among these former programmers
was highly individual; sharing of knowledge only occurred sporadically. As a result, the
evolution of the information systems function did not bring about a significant change in
the dominant occupational routines. The all-prevailing soft-cushion identity of AZ as well
as the dominant engineering conception of the occupation were left untouched.
Because financial resources were not a major issue during its early years - the
department had its own large budget - the demand of and supply for information systems
could grow steadily. Traditionally, this growth in demand was answered by contracting
temporary designers from external software houses. At the beginning of the nineties when
AZ introduced a reform policy in order to increase the amount of service provision, this
admission policy was changed. Because the reform symbolized prosperity, and implied that
the growth in demands for computerized information systems would only increase, it was
decided to hire system designers on a more permanent basis. New entrances were created
and a new group of twenty five system designers in total were hired. These "newcomers "
shared several characteristics which made them strikingly different from the existing group
of designers: the "old-timers ". For example, almost all were in their thirties, and, in
contrast to the old-timers, most new comers had received professional training in
information systems design. During their education and subsequent practical experiences
at other companies, they learned several occupational routines that differed from those
traditionally used at the department. Unlike programming which was perceived as a more
solitary task, system design involved continuous interaction with customers. Formal
documentation of the functional designs, the use of a standard methodology, and the
exchange of experience ("walkthroughs") were considered important professional routines.
Newcomers for example learned that users cannot easily communicate their information
requirements, making constant interaction between designer and users an important part of
ones job. As one of the newcomers remarked:
"Actually we work as sociologists, we constantly try to distillate one reality
out of all the different stories users tell us ... it seems to be pretty difficult
for some people around here "
71
Although because of these characteristics, the newcomes resembled system
designers working at professional software houses, there was one feature that made them
different from them. In contrast to software houses where values such as risk taking, high
income and variety are dominant, AZ' identity espoused values such as security, certainty,
26
and a future which stimulated the newcomers to apply for a position at AZ . It is striking
that these motives did not differ very much from those of the old timers. As a new comer
remarked:
"Look, people decide to work for AZ because it is a company where there
are no intense pressures and where you don't have to work sixty hours a
week to finish your work. On the other hand, your boss doesn't provide you
with a big car, you don't earn a huge salary, and your career won't go that
fast. But on the other hand, you do have a more relaxed working climate,
and more possibilities to work part-time. You see my wife also works and
we have two kids, I can't work sixty hours. Look, I don't work thirty two
hours a week to work eight hours additional during the night".
Consequently, a mixture of social worlds was brought in by this group of
newcomers: they shared with the old-timers their preference for security and safety, while
their occupational knowledge was similar to that present at software houses.
This mixture of social worlds provided a potential opportunity for the existing
group of designers to learn new professional routines from the group of newcomers. After
all, the two groups were not so different from each other as to hinder mutual
communication. Although they had different opinions about the way of doing the job, both
groups felt more at ease with the AZ' soft cushion identity than with a, as "touch"
perceived culture of software houses. However, as will be described below, because of
several learning inefficiencies, the hiring of the newcomers did not result in a change of
the dominant occupational routines.
The staff expansion made it necessary to subdivide the department into several
design sections. The following story is limited to a description of one design section:
36
In should be noted that during that time, the employment opportunities for analysts was relatively
prosperous.
72
"section A". This particular section differed from the other sections in terms of its learning
behavior. Whereas the other sections gradually evolved into more or less professional
groups, section A seemed to have a hard time adapting to more professional work
routines. Although it is hard to determine the exact causes for this difficulty, there are at
least three reasons that may possibly have caused this difference. For one, at section A,
oldtimers surpassed the new comers in number. Whereas other sections were populated by
five to seven system designers of which on average half were newcomers, at section A
twenty designers were employed of which more than half were former AZ programmers.
Furthermore, according to designers of various sections, the users for whom
section A designed systems were more demanding, faced more turnover and the required
systems were more complex to design compared to the users of other sections.
In addition, in contrast to the other sections where relatively young people were
appointed the job of section manager, section A was managed by someone of the old AZ
school The manager shared several characteristics with the general department manager
and the other old-timers. They all were in their forties and fifties, all had received an
engineering education, and because of their years of working at AZ, all had to some extent
internalized the culture of AZ. The years of socialization to the occupational routines
strengthened their world views. According to the managers, the current situation did not
require significant changes.
Section A was coping with a serious problematic relation with the users of the
systems they designed. Users for example complained about the quality of the delivered
systems and the time it took to deliver these systems. Whereas the users pointed at the
section; most section members perceived the users as the wrongdoers. Users could not
specify their information requirements correctly, and when they did, they changed them
constantly.
Although I do not want to doubt the integrity of the complaints of section A, part of
these troubles was a result of learning from past experiences. As mentioned, the oldtimers, including management, used the routines that they had learned during their
occupation as computer programmers. This practice of programming did not require close
contact with users. What is more, as trained engineers they were used to work with
predetermined, well arranged, and fixed specifications. These skills differed importantly
from the skills of the newcomers.
73
Because the oldtimers shared offices and cooperated in projects with the
newcomers, their day to day workpractices made it possible to learn from each other.
However, these interactions enforced only negative sentiments from the side of the
newcomers:
"I know some people of whom I think that given the systems they deliver,
that these people.. they don't belong here anymore. You see, in the past, a
lot of people, people who did not grow up within the age of automation but
who happened to roll into it... they obtained some knowledge and have been
stuck into it. That's it. They haven't changed a bit. And still they persist in
their competence. Really, they're not of much use".
"So you try to improve the communication yourself. But it's.., maybe it's a
bromide, but it has to come from both sides and there are always
colleagues, to put it mildly..., well, we sometimes call them a couple of
snoozers."
Attempts to convince the oldtimers that the section badly needed a change mostly
ended up in frustration and a dissociation from the existing group of designers. Without
being inhibited by management, the old-timers continued doing what they always did.
Some of the newcomers dissociated themselves from the oldtimers by continuing to perform
according to their personal occupational standards. Others dissociated themselves by
becoming more and more discouraged. Since past efforts to make a change at the
organizational level were mostly suppressed or ignored, many newcomers gave up on the
37
power of the dominant coalition .
Only one year after the introduction a reform plan to increase the amount of
services, the decision was made to commercialize. It was time to revitalize the company.
Top information managers of AZ began to discuss the position, function and strategy of the
37
Hirschman (1970) makes a distinction between 'exit, voice and loyalty' when referring to the
strategies people use when faced with organizational decline. The behavior of the newcomers suggest a
fourth, less extreme option than 'exit': dissociation.
74
information systems department. This discussion was also fed by the negative outcomes of
inquiries that were held among the users. For example, it appeared that section A was
often too late in delivering systems, that the systems did not match the specifications of the
users, that the section was considered as operating too bureaucratically, and that
designers were accused of hardly ever visiting their users. Informed by these negative
results, the Director of Information decided to replace the manager of the information
systems design department by a much younger and highly career minded manager who
belonged to the more professional world of information systems. He perceived his job of
"changing the culture" (his own words) as a personal learning experience. Unlike most
designers who identified themselves with the company, this manager identified himself
more with the world of commercial software houses. He propagated the necessity to
become more "cost-aware, client-friendly and commercially minded" and asked for the
participation of the department members in this change-process. Although many designers
of the other sections welcomed the efforts of the new manager, most of the designers at
section A showed a general lack of interest.
This seeming passivity might be due to past experiences. The designers interpreted
the information concerning the change-process that the manager propagated in terms of
their own experiences. They had learned that a manager was the boss who primarily
should command and control subordinates. For example, one of them answered the
question as to why he didn't participate in the change process in the following way:
"[It] doesn't interest me, look that's for the bosses, it's not my job
...I
would like to be good in what I am doing, but I am not paid for other
things, if so they must pay me more".
From years of experience at AZ, these designers had learned not to communicate
informally with bosses, not to see them as equals and not to run the risk of being
perceived as different. Consequently, the new manager's appeal to participate actively in
the change process, for instance by introducing new ideas and by coordinating one of the
many smaller, locally initiated change-projects, was answered by much skepticism.
There was also a history of many reforms which had been initiated but never put
into practice. From this experience of "reforms as a routine" (Brunsson and Olsen 1993)
the oldtimers learned to be highly skeptical about future reform attempts:
75
"... first everything had to be centralized and now everything must be
decentralized, soon if it's all decentralized, everything must be centralized,
it's a strange experience, I must say"
The behavior and attitude of the old-timers frustrated the new manager more and
more. He considered the perceived passivity of the old-timers at section A as a sign of
severe conservatism and adversity to change. In reaction to this, the manager became
more authoritative and oppressive:
"If they cannot change, we can do something about that, if they are not
willing to change, that's something different, we do not need them
anymore".
While pointing to the seriousness of the reform policy of AZ, the new manager told
section A that lay-offs might be considered if they did not change their current behavior.
This only reinforced the ongoing negative learning spiral. For example, the manager's
threat with lay-offs was perceived by the old-timers as a confirmation that a "conspiracy"
was going on among the bosses. The whole reform process was seen as an attempt to get
rid of the oldtimers. As a result, the old-timers felt more or less paralyzed which only
enhanced the manager's perception of the present passivity.
At this point, the research period I had agreed upon with AZ ended. A year after
these events occurred, the department manager moved to a commercial consultancy firm.
The department was significantly reorganized without people being dismissed although
some of the old-timers were appointed to another job within AZ or took an early
retirement.
The organization under study, section A in particular, did not take advantage of the
opportunities it had in changing existing organizational knowledge. Newcomers could have
acted as transmitters of new (occupational) knowledge and as such as initiators of
76
substantial organizational learning.
Institutional knowledge obtained through the education as well as during
experiences at other organizations was introduced into the organization through the
employment of new system-designers. There were opportunities for the externalization and
diffusion of this knowledge among the existing group of designers. Indeed, because
newcomers and old-timers worked together on projects and mostly shared their office with
each other, communication between the two groups was present. However, due to various
internal dynamics that will be discussed in more detail in the coming chapters,
conservatism prevailed. Most newcomers gave up trying to make a difference. By the time
top-management realized the necessity to 'revitalize' the information systems department,
many newcomers were already in the process of unlearning their professional knowledge.
The attempt of the new department manager to change the department and in specific
section A into a user-oriented, commercial organization was answered by much skepticism
and passivity.
In this case story, events appeared rather black and white. To be sure, there were
three other sections within AZ that did not experience such an explicit downward learning
spiral. I have chosen to describe a part of the study that provides the most significant
examples of learning inefficiencies. The causes of these inefficiencies will be discussed in
the coming chapters.
4.4
TRAPS AND OBSTACLES DURING INTERNAL LEARNING
Up until this point, I have discussed the conceptual foundations of internal learning
process. Internal learning serves the purpose of improving existing knowledge through
experience. However, it is not hard to think of traps and obstacles that may hinder the
fulfillment of this purpose. These learning barriers have been depicted by one or more
broken arrows in figure 4.2.
Traps and obstacles occur in situation where organizations learn while assuming
that the circle is closed, when in practice this circle is broken in one or several situations.
In the following, four forms of learning are described that frequently occur in practice.
These forms of learning may obstruct the process of institutionalization. Because
institutionalization, or the process of internal learning, can be considered as the basis of
77
all types of learning that will be discussed in this thesis, the barriers that will be described
in this chapter also apply to feedback learning, learning from others and creative learning.
Audience learning
Organizational knowledge
Individual action
Anarchistic learning
Restrained learning
Individual beliefs
Figure 4.2 Barriers to successful internal learning
- Audience learning
Audience learning occurs when the process of externalization is hampered.
Organizational knowledge construction is not based on individual action although dominant
38
coalition may well think it is (March and Olsen 1976) . The individual does not affect
organizational knowledge - at least not in an unambiguous way.
A learning barrier may be the result of problematic interpretation of individual
action by the 'gatekeepers' of organizational knowledge. In many situations, management
plays an important, sometimes obstructive role as gatekeeper of organizational knowledge.
As part of their task, managers are able to decide what (individual, group and inter
38
Kim (1993) refers to fragmented learning as distinct from audience learning. He describes fragmented
learning in situations where individuals learn but the organization as a whole does not, which to my
knowledge is completely similar to what March and Olsen (1976) mean with audience learning.
78
39
organizational) knowledge becomes organizational knowledge . Management information
systems are one of the prominent instruments which support this translation process
although, as will be argued in chapter nine, this support can be rather problematic.
Because of the influence of gatekeepers, organizational learning can be influenced
by various conservative tendencies. For example, selection of the knowledge can be
influenced by self-referential forces (Huysman et al 1995). Like most human beings,
managers often see what they believe rather than believe what they see. As a result,
managers tend to select information that suits their image of the organization and of
themselves.
Audience learning occurs frequently as a learning barrier and can be the cause of
many conservative tendencies that will be discussed in the next chapter.
At AZ a learning barrier was present during the mutual learning between
newcomers and management of the department. In contrast to the oldtimers, most
newcomers had received a professional education in system-design. Because of this
education and as a result of previous jobs in system-design, these newcomers shared a
professional attitude towards the occupation of system-designer. These shared individual
beliefs guided their actions. For example, they expressed the need to others to
communicate more frequently with users, to make use of a standard design methodology,
to write end-reports, and to introduce "walk-throughs". Management acting as the
dominant coalition of the organizational knowledge, did not value these alternative
'professional' standards which the newcomers introduced. In fact, it did not occur to the
manager, as being an oldtimer himself, that new occupational routines were introduced
within the organization.
The manager
was convinced that there
was enough
communication between him and the group of designers. Although this was probably
correct in the eyes of the managers and the old-timers, the newcomers felt as if their
efforts to change the dominant routines were ignored. This process of 'audience learning'
became crucial in determining lack of significant changes within the department.
- Anarchistic learning
39
This gatekeeping function of management also works the other way around. As translators of the
various information flows, their potential power is impressive (Smircich and Morgan 1982).
79
Whenever the process of internalization is hindered, individual action is not based
40
on internalized organizational knowledge. In such cases I refer to anarchistic learning .
In figure 4.2 this learning process is depicted by the broken arrow between organizational
knowledge and individual beliefs.
Anarchistic learning is a conspicuous form of learning within professional or­
ganizations. In general, the action of professionals such as surgeons and lawyers, is more
driven by professional knowledge acquired during their education than by organizational
knowledge (Abbott 1988).
Learning at AZ provides an example of anarchistic learning. Whereas some
newcomers in time internalized traditional routines, other newcomers could be considered
as "dye-hards". During their socialization at AZ, they deliberately did not internalize
dominant organizational knowledge.
Although I refer to anarchistic learning as a 'learning barrier', to a certain extent,
organizations need some anarchistic behavior in order to learn new ways of thinking and
doing. Individual beliefs that differ from the organizational beliefs can be important
sources for change and innovation. As a group or as a single actor, individuals are at the
center of organizational learning. They often have a front seat while observing the
performances of the organization and that of others (Brown and Duguid 1991). In chapter
seven, when dealing with creative learning, anarchistic learning will be dealt with as being
an important way of promoting the creation of organizational learning.
- Restrained learning
When the process of expression is hampered, I refer to restrained learning.
Restrained learning occurs when members of dominant coalitions think they learn from
individual members although in reality they only learn from individual actions and
expressed beliefs while they ignore underlying personal beliefs. Figure 4.2 portrays this
learning barrier with the broken arrow between individual beliefs and individuals actions.
Situations in which the organization learns without paying adequate attention to private
individual beliefs are not at all exceptional. At least three causes of restrained learning can
40
This concept is similar to 'opportunistic' learning introduced by Kim (1993). He refers to this type of
learning when "organizational actions are taken based on an individual's (or small group of individuals)
action and not on the organization's widely shared mental models" (p. 46). I point at a similar process.
However, whereas Kim is essentially referring to topmanagement acting on their own behalf, I do not
restrict this form of learning as caused by opportunistic reasons.
80
be identified: defensive routines, role-constrained learning, and the power of habits.
Restrained learning can be the result of 'defensive routines' (Argyris 1990).
Organizational members often use routines to mask their vulnerability. This means that
people tend not to be too open about their personal beliefs and tend to act and think in
ways that conform to organizational espoused theories (Janis 1972).
March and Olsen (1976) refer to role-constrained learning when individual learning
has little or no effect on individual behavior as a result of constraints of role-definition,
cultural constraints and standard operating procedures.
Things become even more alarming when individual beliefs do not matter at all; in
this case the organization learns from individual action while this action is triggered by
organizational action instead of personal beliefs. I refer to this type of learning as
'learning by habit'.
Learning by habit is a frequently occurring form of learning during which in­
dividuals learn from the lessons captured in the organizational knowledge repertoire, not
from actual experiences. In fact, this form of learning reflects a simple stimulus-response
model of learning. It deals solely with the retention of experiences; there is no room for
variation of organizational knowledge.
An illustration of this habitualization process has been given by Pauka and
Zunderdorp (1988).
Imagine a cage with monkeys. A banana is hanging on the ceiling of the cage with
a small staircase underneath it. A monkey goes to the staircase to reach the banana. But at
the moment he puts a foot on one of the steps all other monkeys are sprayed. After a
while the same or another monkey tries it again with the same result: again all monkeys
are drenched by spraying. Every monkey that will try to climb the stair hereafter will be
hindered by the others.
Now imagine, we take one monkey out of the cage and replace him with a new
one. This newcomer spots the banana and wants to climb the stairs. To her horror all the
monkeys jump on her neck. After another trial she knows it for sure: whenever you get
on the stairs the others will knock you down. Again, another monkey will be replaced by
a newcomer. And again, the newcomer climbs the stairs and is knocked down. This will
be repeated until every monkey that has experienced the spraying will be replaced. In the
end, no monkey will ever climb the stairs.
81
This story may sound familiar to everyone who has ever been a newcomer in an
organization. Why-questions will often be answered by "just because" answers. Why are
we not allowed to climb the stairs? No one actually knows, it is just because we don't do
that around here. Individual beliefs and individual action are fused together.
Another example of restrained learning can be found among some of the newly
hired system designers at AZ. While they were trying to adapt to guiding routines within
the department, several newcomers began to express feelings of frustration. Although their
personal beliefs were in line with occupational routines that they learned during their
professional education, they perceived it as too demanding to put these private beliefs into
action. This disconnection of individual beliefs and individual action was mainly a result
of the perceived dominant traditional culture. To some of the newcomers, this restrained
learning was the result of previous learning processes, as expressed for example by "I
don't want to stick my neck out anymore". Many had experienced that previous efforts to
introduce alternative routines were ignored or even played down by the oldtimers and
superiors.
- Simultaneous learning
Learning units, such as individuals, groups, teams, and departments, but also
customers, clients, and other stakeholders seldom act as isolated units. Simultaneous
learning by several interacting units can be quite complex and difficult to unravel
(Lounamaa and March 1987). Because of simultaneous learning, units face confusing
experiences. Because learning units are usually unaware of this complexity, intervention
often comes too late. "If one's own actions are embedded in an ecology of the actions of
many others (who are also simultaneously learning and changing), it is not easy to under­
stand what is going on" (Levinthal and March 1993).
The story of AZ provides an example of such inconsistency as a result of confusing
experiences. Because of a policy of reform, twenty-five system designers were hired to
join an existing large group of oldtimers. These newcomers differed significantly from the
oldtimers. For one thing, newcomers shared a rather professional attitude towards the job
of system design learned during their prior professional education. This professional
attitude was almost lacking among the oldtimers. At the time new system designers were
hired, the department gradually became less efficient, at least in the eyes of its users. This
82
inefficiency was due to the very fact that the department lacked a professional standing.
Thus, from an outsider's point of view, one could not imagine a better moment to hire
this new group of professional designers. Unfortunately, the manager's adaptation to his
users was confounded by the adaptation to the old-timers behavior. This confusing
experience in turn influenced the manager's interpretation of the new knowledge that was
brought in by the system designers. The resulting situation prevented the department from
changing into a more professional group of system designers.
4.5
CONCLUDING REMARKS
In this chapter I have described the process of internal learning. Internal learning
must be considered as the skeleton of all other forms of organizational learning discussed
in the following chapters. This idea sharply contrasts with most contemporary ideas on
organizational learning. Many writers on organizational learning perceive organizational
learning as almost synonomous with innovation and change. Learning is then perceived as
thinking (and sometimes acting) in a different way.
In this thesis, the process of organizational learning is considered as many-sided.
Whether organizations learn from feedback information, through imitating others, or
through experimentation, the foundation of all this learning is internal learning. Learning
always involves a reciprocal relationship between externalization of personal (shared)
knowledge and internalization of organizational knowledge.
So far, I have described the process of learning by using a rather static model. It
explains the learning processes of internalization and externalization and ignores
significant changes. Internal learning is an ideal-typical description of a form of learning
that only occurs for example during socialization processes and during the process of
41
sharing individual knowledge, or during the process of learning from past experiences .
Learning occurs within a closed system which leaves only limited room for
41
This latter process has also become known as the 'learning curve' and can be considered as the first
form of organizational learning that has been given explicit attention. It was the US Airforce who discovered
in the 1930's that the direct labor hours needed to complete a production task, decreased significantly as the
total number of times the job was performed increased. This decrease is attributed to the learning that takes
place every time the worker repeats the task. The outcome of this learning is a reduced time and as such
reduced cost per unit.
83
variation. In particular, during internal learning, variation is introduced by hiring
newcomers or by combining existing individual knowledge. In the coming chapters, this
variation will be introduced step by step. Through feedback learning - discussed in the
next chapter - variation will be seen to be a result of organizations adapting to
environmental responses. Through learning from others, variation will be considered as a
result of imitation and cooperation. In the section on creative learning, I will deal
explicitly with the creation of variation.
84
CHAPTER FIVE
FEEDBACK LEARNING
5.1
INTRODUCTION
In the previous chapter, organizational learning processes were treated which are
cut off from external influences. The organization was depicted as a closed system.
Clearly, however, organizations seldom act as closed systems. Organizations are
continuously confronted with external influences to which they adjust.
In this chapter I will delve more deeply into processes of learning from feedback
information derived from the environment. Attention is paid to the general process of
feedback learning; and its functionality will be discussed. This will be followed by a
description of various situations in which learning from the environment may become
problematic. The chapter starts however with a brief review of the literature on
organizations and on their environmental relationship. Given the importance of the
relationship between environment and organizations during feedback learning (as well as
the other two types of learning to be discussed in the coming chapters), it is necessary to
be explicit about the way the organizational environmental relationship will be
approached.
5.2
ORGANIZATIONS AND THEIR ENVIRONMENTS
Relevant literature on the organizational environment has grown so extensive and
heterogeneous that every discussion about organizational environments calls for defining
one's perspective on the concept. This is especially relevant since a preference for one or
more theoretical perspectives on organizational environmental relationships highly
influences further discussions on learning from and with the environment.
Organization studies started to acknowledge the significance of the environment
somewhere in the late 1960's and 1970's. Studies began to treat an organization's environ­
ment as an important determinant of organizational structure and to focus explicit attention
on how variations in exchange relationships led to different patters of organizational
action. The three most important approaches during that time were the contingency
theory, the resource dependence theory, and transaction costs economics. All three
approaches brought the organization's environment into clearer focus.
The contingency theory, pioneered by Woodward (1965), Lawrence and Lorch
(1967) and Thompson (1967) represented an advance over previous theories in portraying
organizations as open systems, dependent on and affected by their environment. These
authors treated adaptive processes as primarily rational and organizations as systems for
transforming inputs into outputs. Organizational structure varied in response to the
complexity and uncertainty of the tasks confronted which determined the nature of
information processing requirements (Galbraith 1973).
The basic premise of the resource dependence theory (Pfeffer & Salancik 1978) is
that organizational behavior can be explained by looking at the organization's context. The
most problematic relation from the organization's perspective is that of dependence on
external social actors. Organizational participants seek to manage these dependencies in a
variety of ways, including bargaining, co-optation, forming trade associations, and
negotiating mergers.
Williamson's (1975, 1985) work in transaction cost economics focusses on the
formation and maintenance of transactions. Williamson contrasts two broad structural
alternatives commonly employed to govern transactions: market systems and hierarchical
structures or organizations. The latter are expected to replace market arrangements as
transactions become more complex, frequent, and uncertain.
One of the serious shortcomings of all three approaches is that they assume
organizations to be essentially rational actors. Many organization theorists contend that
such an assumption is rather problematic (e.g. Brunsson 1985, 1989, March and Olsen
1976, Starbuck 1993). From their study it seems that rationality does not underlie many
actions in organizations. This poses a serious challenge to theories of organizationenvironment relations which presume adaptive rational action on the part of organizations.
Furthermore, these theories downplay or ignore the inter-organizational networks in which
organizations are embedded (Davis and Powell 1992). Network theories (e.g. Burt 1980,
Hakansson 1987, Nohria and Eccles 1991) that came up in the beginning of the eighties,
offer possibilities to challenge this problem.
86
42
Recently, the contributions of the 'new institutionalist' (March and Olsen 1989,
Meyer and Rowan 1977, Powell and DiMaggio 1991, Scott 1987, Zucker 1987) have
extended the perspective of the organizational environment and acknowledge constraints
posed by the environment on organizations. Perhaps the most novel tenet of the
institutional approach is the insistence that organizational environments must be viewed in
cultural as well as in traditional technical and economic terms (Scott 1983). Organizations
and their members are embedded in cultural systems composed of rules, norms, and as­
sumptions which are taken for granted, all of which define the way their world operates.
DiMaggio and Powell (1983) contend that "organizations compete not just for resources
and customers, but for political power and institutional legitimacy, for social as well as
economic fitness" (p. 150). This also entails what Meyer and Rowan (1977) identify as
43
"conforming to rational myths" . The efficiency of these myths are presumed on the
basis of their wide adoption, or their championing by professionals.
An important advantage of the new institutionalism is its recognition of less
rational behavior of organizations in relation to their environments (March and Olsen
1989). Furthermore, it directs attention both to the macro level of legal systems, state
affairs, and profession, and to the micro level of everyday interactions. At the micro
level, institutional practices and beliefs can hinder individual action and individual
cognition. Hence, the new institutional theory is also able to explain the problems
organizations have in change, reorganization and learning.
In line with the institutional perspective, in this and coming chapters cultural
aspects of the environment such as beliefs and institutional practices are seen as having
just as much impact as the more traditional technical and economic aspects. This
knowledge can both constrain the thinking and acting of individuals as well as facilitate it.
It the first situation, rational myths, such as state control, are translated into constraints on
42
The main difference between institutionalism and new institutionalism is that the former is more
oriented towards politics whereas the latter is more sociological oriented. Furthermore, the older variant of
institutionalism is most interested in the local communities in which organizations are embedded. By
contrast, the new institutionalism focuses on non-local environments. Environments are more subtle in their
influence "rather than being co-opted by organizations, they penetrate the organization, creating the lenses
through which actors view the world and the very categories of structures, action and thought" (Powell and
Dimaggio 1991, p 13). See for other differences (Powell and Dimaggio 1991).
43
There are multiple and diverse sources of rational myths, such as public opinion, educational systems,
laws, courts, professions, ideologies, regulatory structures, certifications and accredition bodies, and
governmental requirements (Meyer and Rowan 1977).
87
action. In the latter case, institutional practices and beliefs can be used to construct and
legitimate new courses of action. Professional standards for example are an important
aspect of knowledge derived from institutional environments, with vital potential for
organizational learning.
5.3
THEORIES OF FEEDBACK LEARNING
By feedback learning I refer to learning from ones own experiences through
44
feedback information from the environment . This feedback information can be derived
for example from customers, responding to product quality and price, students responding
to curricula, and citizens responding to social experiments. Hence, feedback learning
requires communication with the environment and can occur through feedback instruments
or through less formalized forms of communications. Examples of feedback instruments
45
are consumer-research, opportunities for public comment, policy-evaluation .
Feedback learning is based on a system-theoretical perspective; it bears some
resemblance to the writings of Argyris and Schon (1978). These authors claim that
organizations
learn from
their failures through feedback
information
from
the
environment. This information can be carried back to correct the action strategies of the
organization. In this instance, the authors talk about 'single loop learning'. Whenever
information is carried back to guiding variables, such as basic norms and values which
direct action strategies, the authors talk about 'double loop learning'. In practice, double
loop learning seems to occur rarely. According to the authors this is because the dominant
organizational theories in use obstruct openness and readiness to change.
This process of single versus double loop learning is pictured in figure 5.1
Although the two concepts have proven to be useful as clarifying concepts, the rare
occurrence of double loop learning has probably more to do with theoretical arguments
which lie behind the concept than with organizational practice an such. Organizational
change is seldom of a discontinuous nature; double loop learning in practice is most often
44
The environment is considered as those people and organizations that do not belong to the subjectively
defined organization though perceived as relevant to this organization.
45
It is not within the scope of this research to analyze these and other mechanisms through which
organizations gather feedback information.
88
Governing variables
À
Action strategies
Consequences
Single loop learning
Double loop learning
Figure 5.1 Single and double loop learning (based on Argyris and Schon 1978)
the result of an accummulation of various single loop learning processes. I will return to
this idea in the next two chapters.
In fact, Argyris and Schon have a rather mechanistic and rational perspective on
organizational learning. They assume that problems of learning such as the difficulties in
achieving double loop learning processes, can be avoided when action is based on
"governing values of Model II: valid information, free and informed choice and internal
commitment" (Argyris and Schon 1978, p. 136). By assuming that these values can be
reached, they stand in almost total opposition to the interpretative and constructivist
approach toward organizational life. Writers within this latter tradition argue that there
cannot be such a thing as valid information, nor is it possible to reach free and informed
choice. Organizational members subjectively interpret information and construct their
environments, while their choices are influenced by significant others.
The idea that organizational learning is often far from being rational has been
acknowledged by March and his colleagues. Differing from Argyris and Schon, they argue
that although organizations may strive to learn in a systematic manner, in practice
organizational life is too unpredictable and complex for learning to occur smoothly and
efficiently (March and Olsen 1976).
To illustrate this idea, they use a model of rational choice. They seek to
89
demonstrate that decision making and learning are seldom founded on rationality. Figure
5.2 depicts this model as an ideal-typical image of learning from the environment.
Organizational knowledge
Individual action
>
f
Environmental reactions
^
Individual beliefs
Figure 5.2 The complete cycle of choice (adapted from March and Olsen 1976)
The model assumes that individuals adapt their beliefs to environmental response.
The change in beliefs or frames of references leads to a change in individual action; this
in turn generates a change in organizational action which corresponds to the response of
the environment. The initial idea of the model is to show that the 'complete cycle of
choice' is full of impediments and will never be closed.
In the following section, I will turn to a discussion of the complexities of feedback
learning when the cycle is broken at one of the four linkages.
Previously in chapter four, problems of learning were treated when the linkage
between individual beliefs and individual actions are broken. Similarly, problems arise
when the linkage between individual actions and organizational knowledge is broken.
Given that feedback learning explicitly deals with learning from the environment, the
present discussion will be restricted to organizational learning while the linkages with
90
46
environmental reaction are broken .
5.4
TRAPS AND OBSTACLES DURING FEEDBACK LEARNING
Feedback learning is susceptible to various obstacles that should be taken into
account as conditions for successful feedback learning. March and Olsen (1976) introdu­
ced four learning barriers while referring to adaptive learning. These learning barriers are
"role constrained learning", "audience learning", "superstitious learning" and "learning
under ambiguity" (see figure 5.3).
Audience learning
Organizational knowledge ^
Superstitious
learning
Environmental reactions
H
Individual action
Restrained learning
Individual beliefs
Learning under ambiguity
Figure 5.3 Incomplete cycle of choice (Adapted from March and Olsen 1976)
Although they all apply to feedback learning, the first two learning barriers are also
present during internal learning as discussed in the previous chapter. Superstitious learning
and learning under ambiguity preeminently belong to situations of learning from feedback
situations.
- Superstitious learning
46
Next to the addition of environmental reaction, the original model of March and Olsen also differs
from the model presented in chapter four (figure 4.2) in that I refer to organizational knowledge while
March and Olsen refer to organizational action. Their is only a small difference between the two concepts.
Organizational action has been considered individual action based on organizational knowledge.
91
Superstitious learning refers to situations in which the organization learns from
information that is wrongly considered as feedback information. Actions and events in the
environment sometimes may have little to do with what the organization does.
Nonetheless, many organizations are very self-centered and perceive these environmental
actions as environmental responses.
"Environmental acts frequently have to be understood in terms of
relationships among events, actors, and structures in the environment, not
as responses to what the organization does. As a result, the same
organizational action will have different responses at different times;
different organizational actions will have the same response" (March and
Olsen 1976, p. 17).
Superstitious learning occurs when environmental action is considered as a reaction
or response to organizational action. Individuals adapt to these reactions which, as a result
of organizational knowledge construction, results in a change of organizational action.
Although it is believed that this action will satisfy the environmental demands, in fact it
will not since it is based on a misconception of the connection between the organization
and its environment.
Avoiding superstitious learning calls for an awareness of the problematic
assumptions within organizations involving their interaction with the environment.
Avoiding superstitious learning thus calls for an awareness of the organizational tendency
to rationalize chance and irrational actions.
"The world of the absurd is sometimes more relevant for our understanding
of organizational phenomena than is the idea of a tight connection between
action and response" (March and Olsen 1976, p. 17)
An example of superstitious learning can be found in the AZ case. Users of the
information systems design department had criticized section A for delivering systems that
were not in line with their expressed information requirements. According to these users,
the section delivered inefficient systems. It was striking that the other three information
systems design sections of the department were not subject to such severe criticism by
their users. A further analysis revealed that the users of the systems developed by section
A were remarkably different from the users of the other sections. For example, most of
92
the users of the criticized section were traditionally confronted with a high turnover of
personnel and various organizational reform attempts. As a result of these changes,
information requirements often became obsolete. Furthermore, according to two
information systems designers who moved from section A to another section, users of the
systems that section A developed, were less willing to communicate with the designers
after they had specified their information needs. Whereas a continuous communication
with the users is nowadays seen as part of the job of designer, the users - dominated by
technical engineers - perceived this as is a sign of incompetence.
Hence, environmental reactions to the actions of section A in the form of severe
criticism from the side of the users was not (only) a result of the malfunctioning of the
section but (also) due to exogenous factors. Nevertheless, the information systems
department took these complaints to heart. Learning from these environmental reactions,
for example by improving their way of working, would only slightly satisfy the expressed
complaints. Perceived malfunctioning of the members of section A may just as well be a
result of the malfunctioning of the users.
- Learning under ambiguity
Learning under ambiguity refers to situations in which it is not clear what happens
within the environment or why it happens, though people impute meaning to certain
environmental events (March and Olsen 1976). It occurs when there is inconsistency
between environmental response and individual beliefs concerning those cues. Instead of
individuals acting on perfect, objective information, individuals construct their own
definition of the situation. The interpretation of environmental responses is often
problematic since environmental actions and events are frequently ambiguous. Moreover,
organizational members have difficulties in observing events, in interpreting them free
from egocentric tendencies, and free from the interpretation offered by others.
Learning under ambiguity has many points in common with audience learning
discussed in chapter four, since both deal with the interpretation of information. Whereas
audience learning refers to internal information, learning under ambiguity refers to
external information.
Figure 5.3 depicts this learning process by a broken arrow between environmental
reactions and individual beliefs.
93
One of the consequences of learning under ambiguity is that organizations tend to
learn in a rather conservative way. Alternative ways of thinking and acting offered by
environmental action and events are frequently by-passed or interpreted in such a way as
to conform to the status quo.
Learning under ambiguity occurred frequently at AZ. An example of this learningbarrier is the attitude-formation of the newly hired 'professional' manager. In his task to
"change the culture" he was confronted with negative reactions of his subordinates. He
translated these environmental reactions in terms of his own frame of reference.
The waiting game played by the oldtimers was understood as a sign of indifference
and even a deliberate attempt to prevent any change, which to the new manager was the
cause of all the problems within section A. Had this manager been more informed about
the previous experiences within the organization, such as the continually expressed urge to
reform without any achievement, and the silent confirmation of their work practice by the
previous manager, he probably would have interpreted this waiting game differently.
5.5
CONCLUDING REMARKS
In this chapter the process of feedback learning has been described. Feedback
learning occurs when organizations learn from their actions through the reactions of the
environment. Feedback learning is susceptible to at least two important conditions that
creates inefficiencies. The first has been referred to as 'superstitious learning', the second
to 'learning under ambiguity'. Superstitious learning is caused by a lack of self-knowledge
and knowledge of the environment. The causes of learning under ambiguity have to do
with difficulties interpreting feedback information. Awareness of its possible occurrence
may help organizations that engage in intentional feedback learning.
In general, feedback learning can result in unintended conservatism. The
organization only learns from its own experiences without keeping an eye on other
environmental events. When organizations solely rely on feedback learning, they will
evolve in a rather ego-centric manner. Only information that is perceived as environmental
reaction is considered as relevant. An organization characterized by an ego-centric
evolution faces the danger of losing its legitimacy and/or competitiveness within a larger
ecology of organizations. In order to avoid this, organizations also learn by adopting
94
knowledge constructed by other organizations. This process of learning from others will
be discussed in the next chapter.
95
CHAPTER SIX
LEARNING FROM OTHERS
6.1
INTRODUCTION
In chapter five it was argued that organizations learn through their interaction with
the environment by adjusting to feedback information. Too much reliance on feedback
learning however may yield conservatism in the long run: the organization learns from its
own experience while ignoring the experiences of other organizations. Learning from
others can be seen as a way to avoid this conservatism, at least to a certain extent.
Learning from others involves the diffusion of external knowledge. External
knowledge is knowledge that is generated by the experience of other organizations.
Organizations capture the experience of other organizations through the transfer of
encoded experience in the form of technologies, codes, procedures, or similar routines
(Levitt and March 1988). It can be obtained through benchmarking, through the use of
gatekeepers and boundary spanners, through the recruitment of individuals, through interorganizational cooperation, mergers, acquisitions, or through less conscious diffusion
processes such as those obtained through institutional forces (Powell and DiMaggio 1991).
Learning from others centers on the acquisition of second-hand experience and has
also been labelled 'vicarious learning' (Huber 1991). When learning from others leads to
becoming more informed about what corporate competitors are doing - and how they do it
- the literature also refers to the term "corporate intelligence" (Porter 1980).
6.2
DIFFUSION OF EXTERNAL KNOWLEDGE
Learning from others calls for an intrusion into the environment. Daft and Weick
(1984) argue that organizations as interpretation systems differ in the extent to which they
actively intrude into the environment.
Passive organizations accept whatever information the environment gives them.
These organizations do not actively search for information within the environment. They
may set up receptors to sense whatever data happen to flow by the organization. They
accept the environment as given.
Other organizations actively search the environment for new knowledge. For
instance, they may hire organizational consultants, technically oriented MBA's; they may
create planning and forecasting departments, or they may engage in intensive
47
networking . In the extreme case of learning by imitation, organizations can even
become inolved in business espionage.
In the literature on innovation, people who act as active receptors have been called
'boundary spanners' (Aldrich and Herker 1977, Leifer and Huber 1977, Tushman and
Scanlan 1981). During the popularity of the contingency theory, many writers pointed to
the importance of boundary spanners for organizational functioning. The scanning of the
environment represents a difficult
organizational problem because people cannot
comprehensively understand the environment (Cyert and March 1963). Learning from
others is rarely a completely rational event. The perceptions of the 'relevant external
knowledge' is crucial for the future functioning of the organization (Daft, Sormune and
Parks 1988). Given that this perception is largely a function of prior related knowledge,
learning from others sometimes results in to path dependency.
There are basically two ways by which external knowledge may diffuse. The first
mechanism is diffusion through organizational alliances or interfirm cooperation. Interfirm
cooperation varies according to the degree in which it is institutionalized. For example,
organizations may learn from each other through the trading of information amongst
(competing) organizations (Hippel 1988), through informal networking (Kreiner and
Schulz 1990) and formal networking (Pennings and Harianto 1992), through collaboration
in R&D organization (Dodgson 1993a), and through mergers and acquisitions.
The other mechanism through which external knowledge may diffuse is via the
process of organizational imitation. Given that the basic process of cooperation is also one
47
I return to this process of informal networking in the next chapter in which also a case story about
informal networking is presented. Although the case provides illustrations of cooperation, the actual process
was more focussed on learning with others than on learning from others and as such is more an illustration
of learning by exploration.
98
of imitation - in fact, inter-organizational cooperation can be considered a process of
mutual imitation - I will restrict the following discussion to learning from others as a
process of organizational imitation.
In the case of pure imitation, organizations - conscious or unconscious - opt for a
'follower' strategy. External knowledge is adopted without a significant adjustment. In
practice, imitation often takes the form of 'creative adoption' in which both the external
knowledge as well as the organization are adjusted in order to find a better match.
According to Levitt and March (1988) the drive to imitate others is often
stimulated by "(p)ressures on organizations to demonstrate that they are acting on
collectively valued purposes in collectively valued ways" (p. 330).
Dimaggio and Powell distinguish three forces
that trigger processes of
48
organizational imitation: coercion, mimicry, and normative pressures . These forces can
be used to explain processes of learning from others and the mechanism for the diffusion
or transfer of external knowledge (Levitt and March 1988).
Coercion refers to a process of diffusion that is more or less imperative.
Organizations become increasingly similar to each other because they are to a certain
extent forced to adopt particular knowledge. This diffusion process can be attributed in
large part both to political influence and to the problem of legitimacy; it often arises
49
without the interference of an intermediary . Examples are governmental regulations,
rules of professional associations, unions or trade associations. Coercive imitation results
from being dependent on other organizations as well as from cultural expectations in the
society within which organizations function. Coercive imitation can be a direct and
48
DiMaggio and Powell (1983) refer to coercive, mimetic, and normative forces as alternative
explanations for the 'iron cage' thesis of Weber. Instead of arguing that organizations become more and
more bureaucratic because this "rational spirit's organizational manifestation was so efficient and powerful a
means of controlling men and women that, once established, the momentum of bureaucratization was
irreversible" (Weber 1922, cited by DiMaggio and Powell, p. 63), Dimaggio and Powell argue that although
organizations do indeed become more and more homogeneous and bureaucracy remains the common
organizational form, this is mainly due to coercive, mimetic, and normative isomorphic change.
While making use of these three concepts, I do not necessarily argue that organizations become
more isomorphic. Rather I use these concepts to explain what reasons organizations have to imitate each
other.
49
Within the theory of mass-communication, this flow of information has been called the 'needle pin
flow of information' (Lowery and De Fleur 1988).
99
explicit imposition of organizational models on dependent organizations. It can occur for
example as a result of legal requirements of the state such as financial reporting
requirements (DiMaggio and Powell 1991). It may also be more subtle and less explicit,
which is for instance the case with writing scientific articles.
Mimicry refers to a process of diffusion that is triggered by an explicit desire to
copy others. A contemporary example of mimicry is organizational bench marking.
Organizations frequently model themselves after other organizations when technologies are
poorly understood (March 1988), when goals are ambiguous or when the environment
creates symbolic uncertainty (Powell and DiMaggio 1991). The copying behavior of Japan
while modeling Western standards, and in turn the Western reaction in modeling Japanese
ways of organizing, are striking examples. Organizations also model themselves after
similar organizations in their field which they perceive to be more legitimate or
successful.
Other organizations
thereby act as dominant models for
receiving
organizations. Brunsson and Olsen (1993) have for example argued that many reform
policies, such as the reform policy of the Swedish Rail, are set up because of such
mimetic processes.
Normative pressures can also be considered as forces that stimulate imitation. This
diffusion process is less conscious than the previously discussed diffusion process and
often stems from professionalization (DiMaggio and Powell 1983). While using the
analogy of an epidemiology of a disease, Levitt and March (1988) refer to normative
pressures as resulting in a two-stage diffusion process involving "the spread of a disease
within a small group by contagion and then by broadcasting from them to the remainder
50
of a population" . Examples are knowledge diffused through educational institutions,
through experts, through selection of personnel, and through trade and popular
publications such as the book of Peters and Waterman (1982) on excellent organizations,
Hammer and Champy (1993) on Business Process Redesign, and Senge (1992) on
organizational learning. These and other books have been used as important recipes to
inform managers how they should act.
50
Within the literature on mass-communication, this process is called "the two-step flow of com­
munication" (Lowery and DeFleur 1988)
100
6.3
ADOPTING EXTERNAL KNOWLEDGE
Up until this point, an impression may have been given that learning from others is
a rather passive activity in that the organization adopts knowledge from outside without
adjusting it to its own needs. Many theorists have taught us already that it is often
impossible only to imitate others without adjusting innovation to the idiosyncratic demands
of the organization, or without adjusting the organization to innovation, or both (Leonard
Barton 1987, Rosenberg 1982, Schumpeter 1934).
Nevertheless, the two theories that explicitly deal with learning from others have
been criticized for not paying full attention to this active aspect of organizational
imitation. Both the literature on institutionalism as well as theories on the diffusion of
innovation are much too focussed on a one-way communication process and as such
provide a rather passive image of organizations.
Institutionalism has the tendency to portray organizations as passive receivers of
institutional practices. Organizations are viewed as static entities, which simply react to
and adapt to the latest trends (Sahlin Anderson 1991). As Powell (1991) argues:
"we need an enhanced understanding of both the sources of heterogeneity in
institutional environments and the processes that generate institutional
change. The literature (on institutional theory MH) suggests a static,
constrained, and over socialized view of organizations" (Powell 1991, p.
183)
The same is true for the literature on diffusion. Although written more than a
decade ago, the following observation of Rogers (1983) still applies to the literature in
general:
"Most past diffusion studies have been based upon a linear model of
communication, defined as the process by which messages are transferred
from a source to a receiver. Such a one-way view of human communication
describes certain types of communication; many kinds of diffusion do indeed
consist of one individual, such as a change agent, informing a potential
adopter about a new idea. But other types of diffusion are more accurately
described by a convergence model, in which communication is defined as a
process in which the participants create and share information with one
101
51
another to reach a mutual understanding" (Rogers 1983, p. xviii) .
In general, learning from others is primarily based on the idea of Schumpeter
(1934) that innovation is merely a combination of technical as well as organizational
innovations; organizations creatively adopt innovations. Or as Sahal insightfully concludes:
"Technical progress is largely a matter of learning by direct experience.
The implication is that there are built-in obstacles to the transfer of
technology, since innovation depends not so much on knowledge imported
from without as it does on experience from within. .. This is not to say that
it cannot be effectively transferred from one organization to the other.
Rather, success in technology transfer hinges upon meticulous alterations in
the design of the chosen technique^ to suit the requirements of differing
production systems. (Sahal 1991, p. 195-197)
Rogers (1983) refers to reinvention when dealing with degree to which an
innovation is changed or modified by a user in the process of its adoption and
implementation. Until the mid-1970's, adopters of innovations were considered to be
passive imitators of innovations, rather than active modifiers and adapters of new ideas.
When diffusion scholars began to analyze the process of implementation, they observed
that quite a lot of reinvention occurred.
Sahlin-Andersen (1991) too has demonstrated that imitating models as a result of
institutional practices, such as coercive, normative and mimetic forces should be seen as a
more active process, rather than the static notion provided by writers within the discipline
of the institutional theory. She refers to the adoption of models as an 'editing' process,
which is the process of translation, re-interpretation or re-invention.
The author argues that the process of editing is restricted by a number of editing
rules. The first set of rules concerns the context. When models are applied in a different
setting from the one where the model has previously been applied or in a setting different
from examples referred to in combination with the model, time- and space-bounded
51
It is striking that, apart from his ideas on re-invention, the work of Rogers can still be criticized for
such a one-way directed perspective on diffusion.
102
features are excluded. Consequently, models that are too bounded in space and time may
not diffuse as easily as those that are possible to edit according to the first set of editing
rules.
A second set of rules concern the formulation and labeling of a model. Models
which attract attention are spreading. Consequently, diffused models tend to be formulated
to attract attention.
A third set of rules concerns the logic of the stories. According to Brunsson
(1989), the dominant logic of organizations is that of rationality. Consequently, models
are formulated according to a problem-solving logic.
I would like to add a fourth rather significant set of editing rules: the model should
not depart too much from existing organizational knowledge.
The ability to recognize the value of new, external information is largely a function
of the organization's level of prior related knowledge (Cohen and Levinthal 1990). This
ability is essential for the success of learning from others and highlights the importance of
previously discussed learning forms. Although too much internal and feedback learning
will result in conservative behavior, the consequences of a lack of these forms of learning
may be even more dramatic. Internal and feedback learning assure the learning of internal
experiences. They guarantee that experiences are translated into organizational knowledge.
Without 'successful' internal and feedback learning, organizations are unable to gain any
benefits from their various activities (March 1991). Through 'successful' internal and
feedback learning organizations may create so-called 'organizational core competencies'
(Prahalad and Hamel 1990).
This case for
incremental knowledge building has its downside. When
organizations are too focussed on the fourth editing rule, they will face the risk of
"unintended conservatism". For example, although the recruitment of individuals can lead
to significant changes in organizational knowledge, most often this practice tends to be
history-dependent since the selection-criteria of the newly hired personnel frequently
reflect the characteristics of the predecessor. March and March (1977) for example found
that school superintendents in Wisconsin were so alike in background as to make further
career advancement random. Kanter (1977) referred to "homosexual reproduction of
management" when addressing the filtering of personnel approaches.
103
Thus, the down side of learning from others - just as that of internal learning and
feedback learning - is the chance of being much too reactive. One of the most frequently
mentioned disadvantages of this conservatism is losing track of ones competitors. Less
often addressed as a down side of conservatism though equally important is the
organizational ignorance of innovative and creative potential within the organization. This
creative potential which is available within organizations as a result of the combination of
individual knowledge, could well produce variance and heterogeneity - the essential
ingredients of knowledge creation (Nonaka 1988). Organizations which seek to attract and
preserve the best knowledge are better off embracing individual creativity and
experimentation. In the next chapter, I will delve more deeply into this process of creative
learning.
6.4
TRAPS AND OBSTACLES DURING LEARNING FROM OTHERS
The most important trap in which organizations may fall during learning from
others is the inefficient capacity to absorb external knowledge. As mentioned, learning
from others cannot take place "ins Blaue hinein". Related organizational knowledge forms
an important part of imitation. Cohen and Levinthal (1990) have coined this function of
organizational knowledge as the 'absorptive capacity' of an organization. Absorptive
capacity is "the ability to recognize the value of new, external information, assimilate it,
and apply it to commercial ends" and is largely a function of prior related knowledge.
With the use of empirical findings about the learning of R&D departments, these authors
argue that the extent to which new ideas can be appropriated and absorbed by an
organization is a function not only of the channels through which ideas spread, but also of
the knowledge capabilities of the receiving organization. Being aware of ones own unique
knowledge through internal learning and feedback learning is thus the first step towards
innovation.
Learning from others can be modelled as a circular process in which through
assimilating external knowledge, organizational knowledge is (re)constructed. This
organizational knowledge in turn directs the environmental scanning process that may
result in the imitation of new knowledge (see figure 6.1).
104
Assimilation
Organizational
knowledge
External
knowledge
Recognition
Figure 6.1 Organizational imitation cycle
52
Imagine the cases in which one of the two arrows is broken . When the arrow
between the construction of organizational knowledge and gathering of new knowledge is
broken, external knowledge is not recognized by the organization as a result of deliberate
search processes. Rather, in such cases introduced external knowledge is for example a
result of unconscious imitation processes or of coercive imitation processes. An example
of this situation can be found in the case story of AZ. The story provides an illustration of
such an 'unmatched' imitation process. The organization under study hired a group of new
system designers who, as a result of their shared educational background, introduced new
'professional' ideas for implementing information systems design. The consequences of
this innovation could have been that through the hiring of new members, the organization
imitated normative professional standards. However, this introduction of new knowledge
was not a result of deliberate action from the part of the department. In fact, the
organization did not even recognize this new knowledge. In the words of Cohen and
Levinthal (1990), the system design department lacked sufficient absorptive capacity to
recognize this new knowledge.
52
As all models, figure 6.1 represents an ideal typical situation of learning from others. Thus, the model
should be seen as an analytical model to assess the learning capacity of organizations.
105
In turn, a disconnection between external knowledge and organizational knowledge
means that the organization is not able to assimilate external knowledge into its existing
way of thinking and doing. This can be the result of not being able to 'edit' the new
knowledge to fit its own idiosyncratic situation. The system design department of AZ for
example, did not adjust its way of thinking and doing with the result that a mismatch
between the newcomers' and the traditional way of thinking and doing remained. Again,
the system design department lacked sufficient absorptive capacity to assimilate this new
knowledge.
Successful assimilation not only depends on related organizational knowledge, it
also requires internal support to implement the external knowledge
53
Implementation of external knowledge requires 'internal networking'. Internal
networking is important to create an awareness within the organization and is needed to
gain the necessary support to implement the innovation. Isolated individual participants
cannot easily contribute to learning from others.
Product champions for example are potential contributors to innovation. A vast
literature on the implementation of innovation has shown that the presence of champions is
an important factor associated with the success of innovations (e.g. Beath 1991,
Burgelman and Sayles 1986, Kanter 1983, Maidique 1980, Schon 1963). This success in
turn, depends heavily on a support group surrounding the champion. As Kanter argues, an
important aspect in the process of innovation is the process of 'coalition building' (Kanter
1988). While most studies emphasize single roles such as the sponsor (Galbraith 1982),
Kanter reveals the importance of a whole coalition, or set of allies. A comparison of over
115 innovations found a set of allies, often peers, behind successful innovations (Kanter
1983). This support group can be seen as the 'dominant coalition' or 'gatekeepers of
organizational knowledge' as discussed in chapter four.
6.5
CONCLUDING REMARKS
Processes of learning from others deal with a specific process of innovation in
which introduction of new knowledge is a result of assimilating external knowledge.
53
106
This condition also applies to creative learning, discussed in chapter seven.
Learning from others is helpful in accumulating external experiences that the organization
itself is unable or unwilling to acquire. As mentioned in section 6.4 the process of
learning from others may be less successful when various conditions are not taken into
account. Furthermore, too much reliance on this form of learning has its own downside.
Past success of imitating other organizations for example, will likely enforce future
patterns of learning since it negatively influences the probability of considering alternative
models to imitate (March 1994).
In addition, relying too much on the experience of others assures that the
organization cannot create its own unique experiences, which are needed to get ahead of
ones competitors. Also, the speed in which technologies are changing nowadays makes it
more and more difficult for organizations to follow and adapt to these chances all the
time. Hence, learning from others should be balanced with creative learning. I return to
this balancing issue in chapter eight.
The next chapter will deal with another aspect of innovation in which the
introduction of new knowledge is a result of experimentation. This 'creative learning' is to
some extent similar to learning from others because the newly introduced knowledge is
almost always a product of a new combination of existing knowledge (Schumpeter 1934).
Furthermore, all preconditions to successful imitation, such as internal networking and
organizational absorptive capacity, also apply to creative learning. Creative learning
departs however from learning from others in that the learning is a result of an internal
drive to create rather than to seek new ideas. It also differs from the learning discussed in
this chapter in that the organization does not so much learn from others as it learns with
others. Through creative learning, organizations may avoid unintended conservatism that
can be a result of internal learning, feedback learning and learning from others.
107
CHAPTER SEVEN
CREATIVE LEARNING
7.1
INTRODUCTION
In this chapter processes of creative learning are described. With creative learning
I refer to innovative processes that are internally initiated within the organization.
54
Through creative learning, organizations create their own 'organizational reality' ,
rather than seek this reality through imitating the environment.
Although approached from different angles, creative learning has received
increased attention in organization and management studies. Senge (1991) for example
refers to generative learning and March (1992) to exploration, as learning processes that
deal with creating new knowledge. The difference with the concept of creative learning
proposed in this chapter and generative learning is that the latter is more focussed on pro­
active ways of learning while creative learning does not necessarily have to be a result of
responding to environmental events. Exploration as a concept introduced in tandem with
its opposite 'exploitation', is a broader concept than creative learning. Exploration consists
of the pursuit of new knowledge, of things that might come to be known (March 1991)
and as such also involves processes of learning from others. Creative learning is a process
of exploration that emphasizes internally triggered variation.
55
Just as learning from others is a form of innovation, so too is creative learning .
Whereas the former is mainly focussed on a search process and the editing of this external
knowledge in order to find a better match with the organizational knowledge, creative
learning deals with an internal drive to create new knowledge. Or in other words, whereas
learning from others deals with organizations that want to imitate, creative learning deals
54
When talking about reality or organizational reality, I refer to a perceived reality, in line with
Thomas' theorem: "If men define situations as real, they are real in their consequences" (Thomas 1928).
55
In line with Rogers (1983) I define innovating as the introduction of an idea, practice, or object that is
perceived as new by the unit of adaption
with organizations that may become models to imitate.
One of the reasons why this type of learning has become popular within the
literature on management and organization, is a growing concern for organizations to gain
an advantage over their competitors. In order to "beat competitors", organizations become
increasingly aware of the necessity to create an organizational core competence (Prahalad
and Hamel 1990). Through learning, organizations are able to build an idiosyncratic
'knowledge base' that rivals find very difficult to imitate.
Furthermore, when learning from others is considered the alternative to creative
learning, there are at least two reasons why organizations may opt for the latter. Rapidly
changing environments make imitating others a difficult enterprise. In addition, as argued
in the previous chapter, imitating requires the matching or 'editing' of external knowledge
to the existing organizational knowledge. This too complicates the process of learning
from others.
Besides mere strategic considerations, organizations that engage in creative
learning are more likely to attract and maintain highly competent and motivated people
than organizations that do not stimulate creativity. After all, organizations that stimulate
(intellectual) creativity may be more attractive to their employees than organizations that
suppress their creative and intellectual potential.
Through a review of literature on ways that could promote internally triggered
variation in organizational knowledge, I will elaborate on various processes of creative
learning in this chapter. Following this review, I will look at the possible occurrence of
traps and obstacles during creative learning. First however, a second intermezzo is
introduced dealing with a case story on creative learning.
Findings presented in this story were obtained from interviews with the EDP
manager and member of the Board of Directors, Mr. Johnson, who can be considered as
the Information Technolog-champion of the particular idea. Six interview sessions took
place from February 1993 to July 1993, providing more than eight hours of recorded
conversation. The interviews were open in the sense that no structured questionnaire was
used during the sessions. The main purpose of this explorative research method was to get
a full account of the learning process that occurred while implementing an innovation.
During the first interview, Johnson gave an account of the organization, his own
110
history and some organizational history. He also described activities that occurred in the
past two years relevant to his 'innovation'. During the five following sessions, he spoke
mainly about activities that occurred within the interval periods and about his plans for the
coming weeks or months. This information therefore consisted partly of recall data - the
first session - and partly of longitudinal data - the next sessions. A year later, several of
the actors mentioned by Johnson, were interviewed in order to increase the reliability of
56
the information .
INTERMEZZO II
CREATING THE MOBILITY PASS AT LEASECO
57
This short case story concerns the very first stages of the process of creating the
Mobility Pass. The idea behind the Mobility Pass was to provide a new service based on
smart-card technology. It was generated by a lease car company "Leasing Co" to
facilitate and control the travel expenses of its client organizations.
Lease Co is a Dutch car leasing company, set up thirty years ago by a banking
corporation. At the time of this research, the Holding of Lease Co had established
fourteen units in twelve European countries, in the United States, and in Australia.
Eighteen hundred people are employed at Lease Co, of which thirty-six at the Holding.
The idea of a Mobility Pass can be seen as an outgrowth of the 'Travel Card'. The
travel card is a credit card that lease-car holders use when filling up their cars with
gasoline. Expenses incurred by the client are then administered at the head office of Lease
Co.
56
The case study demonstrates the importance of conducting research at the very beginning of the
innovation development stage. Innovation theory is predominantly focused on activities that occur after the
decision to innovate has been made (Kimberly, 1981). Questions such as how ideas, problems and needs are
generated often remain unaddressed. As a result, innovations are - mostly implicit - considered a rational
answer to a perceived problem. Instead, research on the process of idea-conceptualization reveals the
dynamic and unorderly process through which innovations arise and even important, others vanish (Rogers
1983). The reality of these post-adoption processes brings to the forefront the social, political and irregular
nature of creating innovations (MacKenzie and Wajcman 1985).
Although the Mobility Pass could have been considered a solution to perceived future societal needs,
it started with a technology driven exploration. If the analysis of the Mobility Pass started at the stage in
which the decision was made to implement the innovation, significant processes such as the importance of
the existing knowledge base, and the failure of most of the projects, were probably neglected.
57
This story is a brief version of a paper presented at ECIS Lisbon (Huysman 1996).
Ill
Because Johnson was the champion of the Travel Card, he gained a lot of
experience with chip-card technology. More important perhaps were contacts that he
established with key players in the world of chip-cards. This know-how, his contacts and
his drive to innovate made him very eager to extend the function of the Travel Card.
Because of his past success with the Travel Card, the Board of Directors left him a free
hand to explore further possibilities for its use.
During that time, Johnson was already in the process of communicating his ideas
with his friends and acquaintances within both the travel industry and the chip-card
industry. After some time, brainstorming about all feasible and unfeasible possibilities
produced a vague and ambiguous vision of a new product that at the same time would
change the general mission of the company. The overall idea was to introduce a chipcard
that clients of Lease Co could use for all kinds of travel expenses and purposes, ranging
from an ID-card used at the airport, to means of payment at car parks and public
transport. This implied that in the future the mission of Lease Co had to change into one
that acknowledged a larger package of services for business-travelers. Leasing cars would
become just one of the various services offered by Lease Co.
In order to communicate his vision to others, Johnson introduced a name for it:
"the Mobility Pass".
Now that the initial vision was born, it was time to explore its possibilities.
Success of the Mobility Pass rested on the cooperation of various actors within the
transport and automotive sector, such as bus companies, railways, parking services etc.
Such cooperation for example involved installation of card-machines in busses, at railway
stations, and car-parks, to scan the pass and register the traveller. Consequently, Johnson
started to contact people active in these sectors. Because of his previous activities with the
Travel Card, many of these people were members of various previously formed network
relations.
Without having a specifically defined concept of the Mobility Pass and without
knowing how his ideas would eventually evolve, Johnson contacted various network
partners in order to explore opportunities. During these discussions, new ideas arose, old
ideas changed, new co-operating partners entered the scene while others left.
From the seemingly inexhaustible opportunities which the vision of the Mobility
112
58
Pass offered, only a few were seriously taken into consideration .
During the period that Johnson and I met, three ideas emerged concerning the
Mobility Pass of which two failed and one reached (as he called it) "the commercial
stage". After the study, this last project also fell through. However, Johnson is still active
in the chip card business, still being convinced that his ideas will finally succeed.
Here is a brief description of the three ideas generated by the Mobility Pass.
One of the first plans that emerged concerned use of the pass to get access to
parking-places in various larger cities in the Netherlands. Johnson started by contacting
the head of a department of parking management in a large city in the Netherlands, who
was a member of Johnson's extensive network. This person seemed to be interested and
introduced Johnson to the company that produced parking meters to discuss the
possibilities of engineering special slots for cards. This company too was interested;
Johnson seemed the right person for them to expand their buyers market. Because they
sold meters to many large cities in Europe, Lease Co could possibly set the standard for
new types of parking machines. Unfortunately, after two months of negotiation, the local
city authorities decided not to invest in the new parking meters which meant that the whole
idea reached a deadlock. Johnson however did not consider it a failure:
"I am as arrogant to say that without the know-how on smart-cards, they cannot
get something off the ground. You need each other, you will not succeed with an
infrastructure only, you also need card owners. (...) I will wait, they will finally
show up".
During the following months, Johnson kept on "stirring in the parking market",
meaning that he tried to create a reputation as the one - and only one - who had the
necessary know-how for smart-cards. By giving interviews and presentations, attending
conferences and workshops, but also by establishing informal contacts, he tried to make
58
Examples of projects that have been taken into consideration before the time of research were the use
of the Mobility Pass to pay for telephones calls, taxies, and trains.
113
himself a known figure in the world of smartcards and the travel industry.
Although not originally anticipated in the early stages of idea generation,
carpooling became another of the many projects within the "mobility puzzle". The idea of
using smart-cards to settle the expenses of carpooling came from a financial investment
bank, Johnson's previous employer. This banking firm had been contacted by a client who
wanted his own project on carpooling with the use of chip-cards, to be financed by the
firm.
Being a former colleague and still a member of Johnson's informal network, the
bank employee in question was knowledgeable about Johnson's experience with chipcards
and his vision of a Mobility Pass. This former colleague then brought him in contact with
the client. Together with a company that promotes carpooling in the Netherlands:
"Carpool Netherlands", they organized a pilot project to explore the possibilities of using
a chipcard to pay for carpooling. However, the pilot study showed that people didn 't mind
having a slot placed in their cars, but what they did mind {due to tax reasons) was to
settle their expenses formally. This finding signalled the end of the carpool project.
Again, enthusiasm of others toward the project was dampened, but Johnson stayed
optimistic:
"I am over-optimistic. For me, I swallow..., a few drinks., and I've got so many
exciting things to do. (..) When I notice that things do not work, I just think of the
story about the thousand ideas of which only two will finally give you profit".
Although this project did not succeed, other ideas emerged out of it as a result of
knowledge diffusion, or the spread of rumors. For example, Carpool Netherlands had its
office in the same building as "Flexlease", a company which had just been set up to
initiate flexible forms of car leasing. Since they were neighbors, the director of Flexlease
heard about Johnson's background and asked him for his assistance in car leasing. In
return he introduced Johnson to a bus company which at that moment was thinking about
starting up introducing shuttlebus services for commuters. The company had not yet
completely formulated the concept; for example it didn't know how to find clients. For this
group, Johnson arrived at precisely the right moment. He could present the necessary
clients (employees) if the company would make use of the mobility pass. Although both
partners had signed a contract, the project finally failed, predominantly because of
114
political reasons.
At the end of my research period, no project had resulted in a decision to commer­
cialize the Mobility Pass. In fact, two years after the research, the whole idea of this pass
is still in a conceptual stage. Enthusiasm for its implementation lessened. Because the
patience of some members of the Board of Directors was at an end, Johnson was asked to
spend more time on other information technological aspects that were considered more
relevant to his field of competence.
In a critique of the prevailing rational myth of innovation as a goal-directed
orderly enterprise, Schon notes:
"In fact, bringing new technology into being is a complex process in which goals
are discovered, determined and modified along the way" (Schon 1967p. 231).
This quotation seems to be very appropriate to characterize the case history of Mr.
Johnson. The whole idea of a mobility-pass emerged out of a lengthy process of creative
learning or knowledge creation.
This example shows clearly the unpredictable nature of creative learning, the
importance of networking, the continuous character of seemingly discontinuous
innovations, and the specific behavior of (IT) champions during creative learning.
It shows that creative learning is centered on the creation of new knowledge albeit
it is not imperative that this newly created knowledge results necessarily in an outcome
such as as organizational innovation. Although the creation of the Mobility Pass was a
case of creative learning, in the end it did not (yet) result in a new product or service.
7.2
INTERNALLY TRIGGERED VARIATION
Whereas during feedback learning and learning from others the introduction of
variety in organizational knowledge takes place from outside, during creative learning
introducing variety takes place from within. During feedback learning and learning from
115
others, variety is brought into play by adaptation to environmental complexities (Ansoff
1965).
The second way of introducing variety in organizational knowledge is much more
diffuse. Internally triggered variation can be introduced through various processes that will
be discussed below.
7.2.1 Enactment processes
Weick (1979) argues that the evolutionary model of 'variation - selection retention', should be adjusted when used to analyze organizational behavior. Instead of
variation, he uses the term enactment to address the more active role of organizational
members in creating the environment which then imposes on them. Besides merely
responding to an independent environment, organizations frequently engage in shaping and
understanding an environment in creative ways.
Enactment is a combination of two things. First, it is "an action of bracketing"
(Weick 1979, p. 130). By bracketing, Weick refers to an action to isolate environmental
changes for closer inspection. These aspects of enactment can be seen as fundamental to
all four learning processes discussed in this thesis. Bracketing leads to a focus on a
specific aspect of the environment. In the AZ case, bracketing occurred for example by
the old timers. Their perception of the environment was based on a selection of
environmental reactions. That is, approval of their superior was the most dominant
environmental reaction; other reactions such as the complaints of their users were not
taken into consideration.
Secondly, enactment refers to active intervention in the environment. In the
specific case of creative learning, this aspect of enactment is particularly relevant.
Enactment "occurs when the actor does something that produces an ecological change,
which change then constrains what he does next, which in turn produces a further
59
ecological change, and so on" (Weick 1979, p. 130) .
It is this idea of active intervention in the environment, that has stimulated Daft
and Weick (1984) to connect enactment with the process of experimenting and innovation.
59
116
Ecological change stands for changes in the enacted environment of the organization.
The authors propose a matrix of four different kinds of organizational processes, each
process characterized by its relationship with its environment (see table 7.1). They name
these relationships "undirected viewing", "conditioned viewing", "discovering" and
"enacting".
Passive intrusiveness
Active intrusiveness
Unanalyzable
Undirected Viewing
Enacting
environment
Constrained
Experimentation,
interpretations.
testing, coercion,
Nonroutine, informal
invent environment.
data. Hunch, rumor,
Learning by doing
chance, opportunities
Analyzable
Conditioned Viewing
Discovering
environment
Interprets within
Formal search.
traditional boundaries.
Questioning, surveys,
Passive detection.
data gathering. Active
Routine, formal data
detection
Table 7.1 Four different kinds of organizations characterized by its relation to its
environment (Taken from Daft and Weick 1984)
Only discovery and enacting as processes of active intrusiveness are of relevance
during innovation. Organizations where discovery predominates are the archetype of a
conventional innovative organization, one which responds - often with great efficiency - to
changes it detects in its environment (Brown and Duguid 1991). Both feedback learning as
well as learning from others are a "discovery" way of learning. By contrast, organizations
where enactment dominates are explorative. Daft and Weick describe enacting
organizations as follows:
"These organizations construct their own environments. They gather
information by trying new behaviors and seeing what happens. They
experiment, test, and stimulate, and they ignore precedent, rules and
traditional expectations (Daft and Weick 1984, p. 288).
117
Thus enactment allows changes to emerge and anticipates its effects instead of
waiting for changed practices to emerge and responding to them.
As mentioned in chapters two and three, the idea of a proactive organization has
stimulated many contemporary writers to "prescribe" a new type of organization: "the
learning organization" (e.g. Garvin 1993, Pedler et al 1991, Senge 1992, Swieringa and
Wierdsma 1990). In this chapter, as in all preceding chapters, the process of
organizational learning is addressed instead of its outcome. Thus, rather than asking what
the outcomes of enactment processes are, I am more interested in what factors produce
60
creative learning processes .
7.2.2 Technology of foolishness
The active intervention aspect of enactment is similar to the notion of 'technology
of foolishness' as opposed to a 'technology of reason' (March 1988). Whereas the latter
refers to the rational well-considered sequence of thinking followed by action, the former
refers to the opposite sequence of action followed by thinking. March argues that
technologies of reason should be complemented with technologies of foolishness. The
notion of a technology of foolishness is a call for playfulness within organizations, "a
deliberate but temporary relaxation of our normal rules so that we can experiment. We
need to play with foolish alternatives and inconsistent possibilities". In order to promote
this technology of foolishness, March (1988) suggests several possibilities. These
possibilities will be reviewed shortly in relation to contemporary
writings on
organizational learning.
- Treat goals as hypotheses.
Organizations should experiment more often with different goals. Contemporary
ideas on organizational learning have introduced the idea of scenario planning and
simulation studies which could stimulate the process of open mind brainstorming about the
consequences of present and alternative goals (DeGeus 1988, Isaacs and Senge 1992,
60
This is also one of the received critics of the theory of Weick. Sandelands and Drazin (1989) state that
the labels of enactment, selection, and retention refer to the outcome of the process stages and not to the
process itself. They conclude that Weick did not keep up to his own task of characterizing organizational
activities in terms of processes. (Sminia 1994).
118
Senge 1992, Stata 1989, Vinnix 1990).
In the Lease Co example, Johnson also treated his goals as hypotheses by keeping
his goals as vague as possible. In fact, he preferred to use the word 'vision' instead of
goals. While testing various scenarios with network partners, an ambiguous vision
gradually became transformed into clearer goals.
- Treat intuition as real.
"When we take intuition more seriously, we could consider alternatives that do not
necessarily rationalize and justify our thinking and acting" (March 1988, p. 263).
Again, Johnson often referred to his intuitive style of working and the support he
got from his colleagues to foster this intuitive thinking and acting. Given that most
organizations are not used to treating intuition seriously, he often met resistance.
"People often want hard figures before they decide, but nothing of the Mobility
Pass was written down, it was just a belief, based on nothing but a certain feeling
about the future".
Although not explicitly referring
to intuition, contemporary
literature on
organizational learning recommends treating 'real' individual belief-systems as more
important than their expressed beliefs. Many authors within the organizational learning
debate advocate open unconditional communication in which theories in action become
public (Argyris 1990, Bohm 1990, Isaacs 1993). These dialogue sessions should be
unconditional in that everything people think or feel should be aired so that not merely
'espoused theories' are shared but real 'theories in use' (Argyris and Schon 1978).
- Treat hypocrisy as a transition.
"A bad man with good intentions may be a man experimenting with the possibility
of becoming good" (March 1988, p. 263). Discouraging experimenting will inhibit
change. Although very true, this suggestion to support a technology of foolishness has to
my knowledge not been taken up by members of the contemporary debate on
organizational learning. Many ambitious "double loop learning" processes such as
reformations, are problematic because they call for a different attitude on the part of the
members, and in specific the managers. These processes could be less problematic if
people are more tolerant of this experimenting.
119
- Treat memory as an enemy.
"If I do not know what I did yesterday or what other people in the organization are
doing today, I can act within the system of reason and still do things that are foolish"
(March 1988, p. 263).
Weick (1979) refers to discrediting organizational knowledge through which
organizational actors treat memory as an enemy: "To doubt is to discredit unequivocal
information, to act decisively is to discredit equivocal information. When things are clear,
doubt; when there is doubt, treat things as if they are clear" (Weick 1979, p. 221).
Discrediting may be a crucial internal source of novelty in the event that this novelty
cannot be borrowed from outsiders.
Discrediting calls for experimenting and enactment. Experimenting is seen by
many writers as essential to organizational learning. Less often mentioned are the
opportunities newcomers provide in that they are unaware of the organizational past. The
AZ case illustrates however that newcomers can only be carriers of new knowledge when
the memory of the organization is not too dominant.
- Treat experience as a theory.
"Personal histories, and national histories, need to be rewritten rather continuously
as a base for the retrospective learning of new self-conceptions" (March 1988, p. 263).
This is a call for self-reflection: in-depth knowledge of the past may assure awareness of
possible obsolete knowledge that still guides action strategies of the organization. This
idea is somewhat similar to the notion of "double loop learning" introduced by Argyris
and Schon (1978) and based on Bateson (1973). Argyris and Schon found their argument
on a system-theoretical perspective on learning. The need for self-reflection is triggered
by an inconsistency or failure. March's suggestion to treat experience as a theory goes
beyond such environmental determinism: "by changing our interpretive concepts now, we
modify what we learned earlier". I will return to this need for self-reflection in chapter
eight.
7.2.3 Diversity
Too much homogeneity creates inertia. When every one is thinking and acting the
same thing, in the same way, interpreting his or her actions in the same way, the
120
probability of learning new knowledge will be low.
Diversity is needed to foster variation in organizational knowledge, different
perceptions of similar situations, enactment of different environments, and the occurrence
of chance, serendipity, and cross fertilization. To put it differently, diversity requires
anarchistic learning. Anarchistic learning has been addressed in chapter four as a form of
learning that may hamper internal learning. During anarchistic learning, individuals do not
act and think as organizational members because the process of
internalizing
organizational learning, or socialization is problematic. Anarchistic learning is not always
negative. In fact, because this process of learning allows for diversity, it may very well
stimulate creative learning.
Diversity is broadly defined in current management literature as including
differences in race, gender, national origin, ethnicity, ability (Milliken and Martins 1996).
Especially in the US, 'managing diversity' is almost considered synonomous with hiring
61
more minorities, and particularly, more female workers . This notion of diversity has
not (yet) entered as folly within European organizations. Diversity is of course not only
linked to a mixture between men and women, black and white, or able and disabled.
Here I approach diversity as differences in worldviews and identity among groups
in relation to other groups.
In general, recruiting and keeping people with different (cultural) backgrounds is
essential to organizations that want to promote creative learning. Through diversity of
personal knowledge, different environments are enacted, new interpretations of the same
situation are given, new ideas emerge, etc.
Another important source of variation is the diversity of roles. In practice,
individuals often play different roles in the organizational setting (Goffman 1959). In
addition to their role as an organizational member, they can play the role of mother,
passenger, client, consumer, peace activist, church member etc. But also within the
61
A feminine way of thinking and acting for example differs from the male way of thinking and acting.
Given that men still dominate most organizations, especially in the higher echelons, a mixture of both female
and male styles of thinking and acting would create diversity. According to Handy (1995) for example,
women rely more typically on intuition than on hard-nosed logic all the time. "Women are less preoccupied
with status, and prefer getting things done. In addition women are quite comfortable at handling three or
four projects or events at the same time. Men tend to want to do things sequentially. Women (and men with
some degree offeminine strain), also recognize that relationships are very important" (Handy 1995, p. 379).
Hiring more women may thus create more variation in organizational knowledge.
121
organizational context, it is possible to think of one individual being member of various
(reference) groups. As Weick puts it "a person does not invest all behavior in a single
group, commitments and interlockings are dispersed among several groups" (Weick 1979,
p. 95).
Inclusions typify the extent to which an actor thinks and acts in a certain social
62
(sub) world , or what the configuration theorists call "social-cognitive configuration"
63
(Bolk 1989, van Dijk 1989, van Dongen 1991, Maas 1988, Veld i n ' t et al 1991). Actors
are always involved in more than one social world. Inclusion in a certain configuration or
social world can be more or less peripheral but is never complete (Veld In 't 1991).
Weick introduced the word partial inclusion to describe partial and incomplete
64
commitment .
Diversity offers both a great opportunity for organizations and presents them with
some important difficulties. On the one hand, more diverse groups and people have the
potential to consider a greater range of perspectives and to generate more high-quality
solutions than less diverse groups (e.g. Watson et al 1993). On the other hand, the greater
the amount of diversity in a group or organization, the less integrated the group will be
and the higher the level of dissatisfaction and turnover (O'Reilly et al 1989). Furthermore,
too much diversity complicates meaningful interaction.
Granovetter (1973), McPerson (1992), as well as Rogers (1983) have warned us
about the paradoxical nature of homogeneity or 'homophily' (Lazarsfeld and Merton 1964)
62
In chapter eight I will elaborate on the possibility of problematic communication due to the existence
of one or more reference groups.
63
The primary difference between social
more focussed on reference groups to which
shared definitions of reality. I prefer the use
emotions, personal attraction, etc whereas
oriented.
64
worlds and social cognitive configuration is that the former is
people want to belong, whereas the latter is more directed at
of social words since this concept includes things as feelings,
the social cognitive configuration is much more cognitive
The social cognitive configuration theorists prefer the term multiple inclusion because, in contrast to
partial inclusion, it means that actor invest their total personality and not only part of it (Maas 1988): "In
patterns of ongoing interaction actors are always included in a plurality of social contexts. Other contexts
are always present in the background. Actors can introduce definitions of reality developed in one
configuration into other configurations they are included in. The process of constructing and reconstructing
definitions of reality is influenced by the multiple inclusions of actors. The existence of multiple inclusions is
an important source of social change." (in t Veld et al 1991, p. 24).
122
within groups. Heterophilous communication may cause cognitive dissonance because an
individual is exposed to messages that are inconsistent with existing beliefs. Homophilous
communication is much more effective (Rogers 1993). But heterophilous communication
has a special informational potential in that it may connect two separate social groups. As
is implied in Granovetter's (1973) theory of "the strength-of-weak-ties", this connection is
especially important in carrying innovative information.
This dynamic can be illustrated by the AZ case. Newcomers and oldtimers were
members of different social worlds: the world of the professionals versus the soft-cushion
world of AZ. Because these two worlds were so different from each other, no meaningful
communication occurred: both groups did not (want to) understand each other.
Thus, too much diversity is as dysfunctional for change as is too much similarity.
Besides embracing differences, diversity may be enhanced through serendipity and
cross-fertilization.
As Koestler remarks in his essay on the act of creation:
"The most important feature of original experimental thinking is the discovery of
overlap and agreement where formerly only isolation and difference was
recognized" (Koestler 1964, p. 232)
There are many ways through which serendipity and cross-fertilization enter the
organization. For example, attendance at conferences where the subject is at first sight of
little relevance, may be a fruitful explorative activity. The same goes for hiring external
guests whose expertise differs from the major expertise found in the organization. The use
of Internet-facilities, especially World Wide Web and the bulletin board facilities, is a
good example of explorative information systems that promote serendipitous findings and
the connection of separate disciplines. I will return to the use of information systems to
promote creativity in chapter nine.
7.2.4 Informal networking
During the process of creating the Mobility Pass, a lot of informal interactions
across various organizations occurred. Johnson had ties to multiple networks which he
used for various reasons. For one, he needed the collaboration of actors within the travel
industry in order to solve the "mobility-puzzle". Another important reason to network was
123
to learn with others. Partners were more often used as brainstorm or sparring partners
than as trading partners. Network ties were used to refine loose ideas, to get inspiration
for new ideas, to change existing ideas, and to get in contact with others with whom he
had no ties at that point. Thus, communication with people outside the organization
65
facilitated creative learning .
Inter-organizational networking can play a vital role during creative learning. By
tapping 'external' knowledge, ideas and experiences get blended which can lead to new
linkages and new insights (Pennings and Harianto 1992). When networking occurs in an
unstructured way and partners move freely along various sources of information,
unanticipated sources of knowledge may arise. This happened at Lease Co, where Johnson
- like all innovation champions (Kanter, 1983) - promoted his idea vigorously through all
kinds of informal processes. As a result, pieces of information flew to places where they
were neither asked for nor, according to conventional wisdom, relevant. Most projects, of
which only one has been described in this paper, were not planned for or even thought of;
the information just moved in the form of hearsay to unanticipated places.
Similar processes were observed by Kreiner and Schultz (1993) who studied the
informal collaboration of sixteen R&D departments within the Danish biotech community.
To them, networking seems to embody a "technology of foolishness" (March 1988).
Instead of the rational sequence of action following thinking, networking often involves
the opposite. In a similar vein, Johnson's networking efforts can be seen as rational use of
a technology of foolishness. His aggressive networking was a result of his belief that
contacts, although not useful in the short run, can always prove to be of value in
subsequent stages or projects, and that through networking he could establish a certain
reputation. Whereas the motivations that laid behind this networking were rational, the
actual process of networking was more one of 'thinking follows acting':
"I keep on beating the drum loud and clear. Very often, people have needs, do not
65
It is striking to note that although Lease Co, with car leasing as its main product, is a competitor in
the eyes of other actors in the public transport sector, they did not perceive each other as rivals when
exploring opportunities was concerned. Clearly, this observation departs from the prevalent aggressive notion
of competitive struggle (Porter 1980).
The use of actual and potential competitors in order to exchange information concerning novel
designs or product ideas is not that revolutionary as it seems to be. In fact, a century ago, "collective
invention" was probably the most important source of innovation (Allen 1983).
124
know precisely how it sticks together and go to seminars or so, seeking what can
be done about it. It is a kind of cattle trade in knowledge. Most people do not read
magazines like: I have a problem, let me see if there is some intellectual who has a
solution. They feel betrayed by their suppliers, suppliers always come with
something new. They don't want that any more, the push market is over."
7.2.5 Risk-taking
Another important ingredient of creative learning is the nourishing of risk taking
and risk takers. Creative learning depends on the relation between performance and
aspirations (March and Shapira 1987). Individuals tend to act in a more risk averse
manner when they are above their aspiration levels than when they are below them. When
operating below the aspiration level, individuals seem to increase risk taking as they fall
further below the target. Risk averseness sets in at the moment individuals reach a
perceived 'survival point', when they find themselves in situations in which performances
are very much below the aspiration level. Above the aspiration level, risk taking seems to
rise slowly with success. Thus failure (until survival is in question) and substantial success
induces risk taking (Levinthal and March 1994).
It is important to note that risk taking is not only associated with failure and
success, but also depends on the role of beliefs of individuals. Personal aspirations are
most decisive for risk taking. Aspirations can be suggested by 'significant others' that act
as reference groups, such as other organizations, superiors, etc. Aspirations are also
developed through personal experiences. In such case, risk taking is not so much a result
of perceived external threats or other problematic situations. Rather, past individual
successes often triggers future risk taking behavior. This is primarily because successful
people tend to overestimate the contribution of their ability to their success and to
underestimate the contributions of risk taking and chance (Levintal and March 1994) Such
risk underestimation reinforces illusions of control. In other words, those individuals who
have experienced past successes tend to engage in more risky behavior than they would if
they understood the odds.
Again, the Lease Co case provides an example of this risk taking. Confirming the
characteristics of IT champions (Heng et al 1994), Johnson liked to take risks and to a
125
certain extent ignore formal routines.
" ƒ have my own vision about how things ought to go. They [Board of Directors,
author] don't like that at all. It's accepted now, after five and a half years"... "I
compare my self with the joker of the Middle Ages. In fact you may say almost
everything; if you are right, it will be accepted, if you are wrong, everyone roars".
There is an increased attention within the literature on innovation toward the
importance of risk takers for the broaching of innovative avenues. These risk takers have
been labelled 'product champions' or 'innovation champions'(Burgelman and Sayles 1986,
Kanter 1983, Maidique 1980, Schon 1963). These individuals share a certain thirst for
tension and excitement that stimulates them to introduce new ideas.
We are probably all familiar with those people who have a sort of innate drive to
search for or to create risky and dangerous situations. As Johnson remarked:
"You always need food also for your brains, only it depends on the way you
are brought up, how to deal with the food. When you belong to the ninetyfive percent of the population that were never allowed to take risks, you will
never be creative. When you are within the five percent who is allowed to
behave a-socially, than you will manage. When you behave as such within
your professional life people react kind of jealous but they forget that I am
continuously walking on a wobbly edge of being kicked out because there is
no result"
As the process of idea generation of the Mobility Pass illustrated, most projects - if
not all - do not succeed. In addition to the two projects described above, a lot of other
ideas failed. However, it is incorrect to deem the efforts put in the failed projects as
waste. Creative learning calls for embracing risk taking while failures should be regarded
as part of chosen strategy. The generation of an innovation must therefore be seen as a
learning process and the experiences learned by trial and error form part of the success of
the innovation. This implies that organizations who promote creative learning should
consider this learning by doing as an economic investment (Arrow 1962). In addition to
the positive effect of knowledge building through learning by doing, contacts and
friendship forged in the preparing phase of "failed" projects, may prove useful for
126
subsequent projects. To invest in this creative learning capacity, organizations need to
change the prevalent assumption of "fighting for survival" into an assumption of learning
with others.
7.3
TRAPS AND OBSTACLES DURING CREATIVE LEARNING
As I will discuss more thoroughly in the next chapter, creative learning stands on
the shoulder of the other three forms of learning. Hence, conditions described in the
previous chapters also apply to this form of learning. But organizations may also face
inefficiencies during learning that are salient to this particular type of learning.
Although creative learning evokes images of inventing, creativity and originality,
in practice its success depends considerably on prior related knowledge. In the previous
chapter, the importance of the organization's absorptive capacity in order to innovate has
already been discussed. In the case of creative learning, this absorptive capacity can best
be described by Pasteur's famous expression "chance favors the prepared mind". Past
experiences represent a rich source from which creativity may flow. New ideas do not just
fall from trees, they are always based to a certain extent on prior knowledge.
The story of the Mobility Pass illustrates the importance of prior experience. The
success of a previously introduced product, the Travel Card, triggered the Holding to
think of ways to extend its function. This same story demonstrates the importance of
existing organizational knowledge and highlights the particular nature of seemingly
discontinuous (McKee 1992), radical (Burgelman and Sayles 1986) or big bang (Gluck
1985) innovations, as well as the process of double loop learning.
The process of creating the Mobility Pass can be considered a cumulative
technological experience, although it would be tempting to consider the innovation as
discontinuous in case it had been analyzed ex poste. If Lease Co had had no know-how of
smart-cards, the company probably would not have thought of introducing the Mobility
Pass in the first place. To summarize the argument in the words of Cohen and Levinthal
(1990), Lease Co had enough "absorptive capacity" to work out the idea of the Mobility
Pass.
Hence, creative learning calls for exploring the future while exploiting the past.
127
However, by contrast, organizations should not rely too much on past experiences.
After all, the purpose of creative learning is to produce radical new knowledge. In order
to reach this goal, actors engaging in experimenting should be cut off from the rest of the
organization in order to be receptive to totally irrelevant ideas. This need to be isolated
from the rest of the organization is also essential when we realize that the products of
creative learning will only become feasible after a relatively long period of time. Given
that organizations are predominantly short-term oriented, creative actors would likely be
confronted with impatience on the part of management (March 1994).
The Lease Co case provides an illustration for such a situation. After almost four
years searching for opportunities to extent the function of the Mobility Pass, the Board of
Directors of Lease Co realized that Johnson's efforts would not yield any fruitful
outcomes in the near future. One of the members of the Board talked about a "waste of
time" and thought the time had come for Johnson to focus on the "regular activities as an
EDP manager instead of playing around". In fact, one year after the study Johnson was
asked to concentrate more on the actual internal affairs within the company.
Too much focus on creative learning on the other hand will also produce
inefficiencies. Through excessive creativity, organizations may fall in a 'failure trap'
(Levinthal and March 1994). Because most new ideas are bad ones, on average creative
learning will lead to disappointment. And because new ideas require time to realize their
value, organizations often shift to new alternatives before they develop the expertise
necessary to exploit old ones. Consequently, creativity often leads to perceived failure
which leads to new searches, which leads to failures, and so on (Cyert and March 1963,
March and Simon 1993). In order to interrupt this cycle, creative learning should be
balanced with other types of learning. In the next chapter, I will return to this issue of
balancing learning.
7.4
CONCLUDING REMARKS
In this chapter I discussed processes of creative learning. Of all forms of learning
processes that have been treated in this thesis, creative learning is most closely related to
learning that has been labeled "generative learning" (Senge 1991), or "proactive learning"
(Miles and Randolph 1980). These theories portray significant changes at the level of the
128
organization as an outcome of learning.
Different from these writings on organizational learning, I do not want to argue
that creative learning is the best way of learning. Rather, creative learning is based on all
other forms of learning described previously. For example, without learning by imitation,
creative learning is hard to imagine. As Schumpeter (1934) argued, innovation is always
the result of "Neue Kombinationen". Furthermore, organizations engaging in creativity
without paying attention to existing knowledge within the larger environment face the
danger of losing track.
In the coming chapter I will argue that the possibility of successful outcomes of
creative learning, that is an increase in the breadth of organizational knowledge, depends
to a great extent on the success of the other forms of learning.
129
PART THREE
IMPLICATIONS
In this third part of the thesis, two chapters are devoted to the implications of the
ideas put forward in the previous chapters. Although it is not the intention to provide
standard recipes or guidelines for organizational practitioners, some words may be said
concerning the practice of organizational learning. A distinction is made between
implications for organizational practitioners to promote successful outcomes of learning,
and implications for the information systems discipline. By organizational practitioners, I
refer to those people who are the most capable of purposefully introducing, changing or
suppressing organizational learning processes. In particular, 'human resource managers',
leaders, organization- and management consultants, but also organizational researchers,
are referred to. To be sure, implications for the information systems discipline are of
equal interest to these and other practitioners.
The implications for organizational practitioners addresses the question how to
strive for
'successful' outcomes of organizational learning processes. Given that
organizational learning processes are often inefficient, the question will be approached by
addressing the causes of imperfect learning and attempting to show how organizations can
try to avoid these causes. It will be argued that organizations strive for successful
outcomes of learning by circumventing instances offocussed learning and by balancing the
four types of learning.
In chapter nine, the implications for the information systems discipline are
addressed. The chapter deals with reviewing the role of information systems during several
information intensive processes that characterizes learning. These processes are:
knowledge externalization, knowledge objectivation, knowledge internalization, information
selection, information interpretation, and idea generation.
CHAPTER EIGHT
STRIVING FOR SUCCESSFUL LEARNING
8.1
INTRODUCTION
Up until this point, processes of organizational learning have been discussed
theoretically without explicit reference to the possibility of achieving successful outcomes
of learning, such as improvement, innovation, and intelligence. How to strive for
successful learning is a particularly relevant issue when we realize that organizational
learning processes are prone to various inefficiencies.
In this chapter, I will elaborate on the issue of improving learning capabilities. The
general idea behind this attempt is that in order for organizations to improve their learning
capacity, it is necessary to create awareness of possible learning imperfections. As soon as
this awareness is brought about, there are various ways to avoid the occurrence of these
imperfections.
It is tempting to solve problems by presenting an ideal type of learning. Such an
attempt comes close to presenting normative descriptions such as what a learning
organization should look like. This may be criticized however for being much too biased
towards one particular type of organization, within one particular context. To be sure,
there are no twin organizations. Prescriptions as to what learning organizations should
look like may be applicable to one organization but not to others. Because the
organizational past history, the idiosyncracies of organizational members, organizational
institutional contexts - all these and more - affect the process of learning, time and context
are important variables that may affect the learning itself. It is for example conceivable
that the learning of organizations during the industrial revolution results in a different
form of learning from for example organizations during the 'information revolution'
(Beninger 1986) or the 'knowledge society' (Drucker 1988). Furthermore, an increase in
amount of organizations within an organizational field assures a more 'aggressive' style of
learning than is the case with an organization in a sparsely populated organizational field.
Likewise, organizations in Sweden with their tradition of work councils will likely
engender a different style of learning than do organizations in Belgium where lines of
66
authority are also informally kept in touch (Hofstede 1980) . The nature of the firm also
influences the best way of learning. One-man businesses learn differently than multi­
national corporations.
Hence, since I do not believe there is one best way of learning, it is not my aim to
present normative prescriptions. Another, more fruitful option that still acknowledges the
idiosyncracies which characterize organizational life, is to present descriptions about when
and how learning processes may or may not yield fruitful outcomes. From these
descriptions some generic implications may be derived that could be of help whenever
organizations want to engineer their learning process.
In the previous chapters it has been argued that organizational learning does not
always result in outcomes such as improvement or intelligence. In this chapter I will
consider how organizations can avoid the occurrence of these and other learning defects.
This will be done by first addressing the causes of the traps and obstacles to learning that
have been discussed in chapters four and five when internal learning and feedback
learning were described. All these inefficiencies have to do with focussed selection,
interpretation, and use of information. Given that the other two types of learning: learning
from others and creative learning, are essentially information intensive processes, this so
called "focussed learning" may also influence these two types of learning. After reviewing
the various tendencies that cause focussed learning, I will discuss possible ways to avoid
its occurrence.
Avoiding focussed learning is not the only way to improve learning capacity in an
organization. In order to promote successful learning, every type of learning should make
use of other types of learning. In section 8.3, I will go into every possible combination of
types of learning and how this may stimulate improvement and intelligence. In an
appendix at the end of the thesis, a general checklist is given that can be used to assess
organizational learning capacity. This checklist is based on arguments put forward in this
chapter.
66
The differences between organizations in Japan and the US in terms of learning have been addressed
frequently (Nonaka 1990, Nonaka and Johansson 1985)
134
8.2
FOCUSSED LEARNING
The traps and obstacles discussed in chapters four and five are a result of
selectively searching, interpreting, and using information. There are various tendencies
that may cause this focussed learning. These tendencies hinder the introduction of variety
or diversity of information and knowledge within the organization and as such result in
path dependency. Path dependency is one of the most important perils of organizational
learning. It is often a result of unintended conservatism. Standard evolution theory already
taught us that evolution without variation will sow the seeds of destruction. The
organization continues doing what it always did without looking for alternative courses of
.67
1
action
Specialization
Audience learning
Physical and
cultural conditions
Anarchistic learning
Reference groups
Focussed learning
Self-reference
Restrained learning
Simultaneous learning
Learning under ambiquity
Hidden learning
Superstitious learning
Figure 8.1 Causes and effects of focussed learning
Below I will discuss five tendencies that cause focussed learning: specialization,
physical and cultural conditions, self-reference, reference groups, and hidden learning.
Figure 8.1 depicts the relationship between the causes and effects of focussed learning.
Secion 8.2.6 deals with other possibilities than sheer awareness of its occurance, for
avoiding focussed learning.
67
Path dependency may also be the result of unintended chaos. This may occur when there is an
overreliance on experimentation without capitalizing on the experience gained through experimenting.
135
8.2.1 Specialization
Because of specialization among individual employees and among organizational
units - departmentalization
individuals perceive their environments differently. This
selective exposure to environmental stimuli affects the information that various members
receive: "Salesmen live in an environment of customers; company treasurers in an
environment of bankers; each sees a quite distinct part of the world" (March and Simon
1993, p. 175). Through the division of work "perceptions of the environment are biased
even before they experience the filtering action of the frame of reference of the perceiver"
(March and Simon 1993, p 174).
Selective exposure induced by specialization assures the persistence of sub-goals
and the interpretation of environmental responses in terms of these sub-goals. Because
units or individuals think in pieces instead of wholes, the possibility of perceiving environ­
mental action which is less relevant to ones own segment though perhaps relevant to
others', is kept to a minimum (Senge 1992). Railway drivers sporadically meet managers
(Edel 1996); consultants are asked to plan a day to meet their fellow colleagues (Peters
1992); salesmen seldom talk to marketing people (Senge 1992).
System theorists have devoted considerable attention to the problem of selective
attention. For system theorists, a source of poor performance and organizational failure is
often to be found in the limited cognitive skills and capabilities of individuals when set
against the complexity of the systems they are addressing (Forester 1961, Galbraith 1973,
Perrow 1986, Senge 1992, Simon 1977). Most if not all of these system thinkers perceive
selective exposure as a result of specialization. In addition to system-theoretical
considerations, cultural and physical conditions may also produce selective exposure.
8.2.2 Physical and cultural conditions
Most communication in organizations is informal and takes place for example in
the coffee corner, during lunch hours, on the way to the office, or when popping in at the
office of colleagues. Physical as well as cultural conditions may however block the
occurrence of such unexpected encounters. At AZ for example, it was one of the
unspoken rules not to talk informally with each other, at least not when the boss was
walking around. Consequently, most informal communication occurred behind closed
136
doors and more importantly was almost restricted to those people who shared an office
with each other. Aside from the fact that these people only occasionally talked informally
to their fellow-colleagues, they seemed to strictly avoid contact with superiors. Again, this
was one of the unwritten rules at the department: managers only talk formally with their
subordinates. In fact, old-timers perceived some newcomers who had not internalized this
rule, as always on the make, doing everything to get higher up in the hierarchy.
There were also structural hindrances at AZ to informal communication. Offices
were located on two sides of a very large and narrow corridor. Except for the lunch room
downstairs, there was no possibility for employees to meet and talk to each other
informally. Although there was indeed a common coffee-machine, this machine was
located in front of the office of the manager. Given the hierarchical culture at the
department, this was not exactly an ideal place for an exchange of ideas and opinions.
Hence, physical and cultural conditions may limit the scope of information to which one is
exposed.
8.2.3 Reference groups
Selective attention and interpretation of information can also result from ones
68
membership in different social groups . Membership in social groups may influence how
an organizational member determines what information to interpret, how to interpret it,
and how to use it.
In this thesis, the concept of reference groups is used when dealing with social
groups. Reference groups are the source of a person's aspirations and evaluative beliefs.
They can be seen as groups whose perspective is used as a frame of reference by the actor
(Shibutani 1955). This notion stems from symbolic interactionist thinking with Herbert
Mead as its intellectual father. Symbolic interactionists argue that the self, that is ones
own identity, emerges through the process of social interaction with others. People
imaginatively internalize attitudes of the reference group to which they want to belong.
People, according to Mead (1934), tend to "take the role of the generalized others",
68
Here we meet social psychological theory. Social psychology is the study of how people think, feel,
act and learn when they are with others, or when they have others in mind (Hosking and Morley 1991).
137
meaning that each person approaches his world from the standpoint of the shared
perspectives or frame of reference of the group.
An important feature of reference groups is that these groups are not necessarily
bound by geography or formal membership "but by the limits of effective communication"
(Shibatuni 1955, p. 566). All kinds of groupings, with great variation in size,
composition, and structure, may become reference groups. Their formation may be
attributed to some social category - a social class, a community, an ethnic group, or to
groups in which people participate directly - a work team, an occupation. A reference
group may also be imaginary, as in the case of artists who are "born ahead of their
times".
Groups determine the expectations and interpretations of people's activities. The
amount of cohesiveness in these social groups is largely a function of the degree of
corresponding interpretive schemes. Strong group cohesion creates a situation of
'cognitive consensuality' which could be defined as "a reasonable amount of implicit
agreement among organization members as to the appropriate meaning of information or
events" (Finney and Mitroff 1986 p. 320). Cognitive consensuality may foster a group
atmosphere in which people take too much on trust and suppress their personal doubts
about what is being said. Janis has called this 'group think'. People are likely to persuade
themselves that their "misgivings are not relevant" and that "the benefit of any doubt
should be given to the group consensus" (Janis, 1972, p. 201) In the case where
individuals are hindered from putting their beliefs into action, a kind of group think may
emerge which promotes 'restrained learning' as discussed in chapter four. Group think
may also influence learning under ambiguity when people attend to information and
interpret this information according to the group norms.
The AZ story illustrates that reference groups may seriously block introduction of
knowledge that does not match the general frame of reference of the group. System
designers either belonged to the AZ culture and tradition or to the professional world of
software houses. The latter group of designers, at least at the beginning, reacted to actions
of the old-timers by pointing to other - in their eyes - more improved ways of working.
These efforts had no result, mainly because the gap between the two social worlds was too
big.
138
8.2.4 Self-reference
Another important source that may influence learning is self-reference. The
concept of self-reference is derived from the theory of autopoiesis (Maturana and Varela
1980) and has been used by Morgan (1986) as one of his metaphors to analyze
organizational behavior.
The term 'autopoiesis' stems from the Greek avroa (self) and woieiv (to make).
The theory is a new approach to systems theory since it challenges the traditional
distinction between a system and its environment. It also provides a new perspective on
the evolution of living systems. One of the principle features of the theory is the concept
of 'self-reference'. Self-reference means that systems make reference only to themselves.
That is, elements of the system interact only with other elements of the system. The
system cannot interact with elements outside itself because these are not specified. Thus, a
system's interaction with its "environment" is really a reflection and part of its own
organization. It only perceives the environment as a projection of its own identity.
Because of this self-reference, the evolution of living systems is characterized by selfproduction.
In his book "Images of Organizations" (1986), Morgan discusses the theory of
autopoiesis from an organization theoretical point of view. He uses the theory of
autopoiesis for his "flux and transformation" metaphor, which deals with the logic of
transformation
and change for
the basic dynamics that generate
and
sustain
69
organization . This metaphor provides a totally different idea about the distinction
between organization and its environment.
Traditionally, organization theory treats environments as exogenous, whereas the
organization is supposed to act adaptively in order to respond to environmental signals.
Organizations are considered open systems continuously trying to adapt to changing
environments. The theory of autopoiesis, by contrast, provides a picture of organizations
as being closed systems. This closedness should not however be taken too literally.
69
The use of this metaphor is appropriate, but not the only possibility. Autopoiesis does deal with
change and reproduction of systems, it also deals with maintaining its own identity and psychology.
Consequently, the theory of autopoiesis might be of use for Morgan's "psychic prison" metaphor as well.
Because the theory originates from biology, it would also be of use for the "organism" metaphor.
139
Organizations do have environments, but their relations with environments are internally
determined. This notion of endogenous environments is similar to the idea that
organizations enact their environment: environments are socially constructed and come
into being by giving meaning to it (Weick 1979). Although enactment is usually
considered a process of development, it "encourages us to view organizational enactments
as part of self-referential process through which an organization attempts to tie down and
70
reproduce its identity" (Morgan, 1986 p. 241) .
Although not discussed by Morgan (1986), the theory of autopoiesis has also
promising potentials when used as a metaphor to analyze organizational learning processes
(Huysman et al 1995). Used in this fashion, a self-referential image of learning arises,
implying that organizational learning is biased by the existing organizational identity. This
identity operates as a filter by defining what history and "outside" events are considered
relevant in order to draw conclusions from them. Because of this 'structural coupling',
learning is self-referential: the object of learning is related to what is already known or at
least understandable. This self-referential learning can also be seen as what Schon (1971)
described as 'dynamic conservatism': a tendency to fight to remain the same.
There is some overlap with the existence of reference groups as sources for
learning under ambiguity. For example, both deal with the dominance of (group) identity
in attending to environmental demands. However, whereas the concept of reference groups
is directed at significant others, self reference is related to ones own identity.
Furthermore, the theory of autopoiesis used as a metaphor to explain focussed learning
argues that the identity of the group or organization is important not so much in
interpreting the environment but in creating the environment. When self-reference is
considered a source of learning under ambiguity, it is assumed that the individual stays
within his or her existing frame of reference and chooses environmental responses that
confirm this frame of reference. The pictures of environmental responses formed by each
employee are representations of reality created through self-referential mental processes,
rather than objective pictures of 'reality itself (Westenholz 1993). This does not
necessarily mean that there is no existing reality, rather it implies that we construct
70
Enactment can be both creative as well as conservative. I will turn to the first aspect of enactment
when discussing learning by exploration. For now, I will address the conservative side of enactment is
addressed.
140
pictures of the world that decisively shape our world view. People create these pictures to
defend their identity but at the same time "these pictures become a prison into which we
are locked, so that we cannot view the world afresh" (Westenholz 1993, p. 39).
Hence, organizational learning may take place within the limits of the existing
mind-sets. Because these limits are the very means by which the information is interpreted
71
as meaningful, individuals will not be capable of challenging these limits . In other
words, organizations tend to learn from an ego of themselves. Superstitious learning is a
conspicuous result of egocentrism. Because the organization perceives itself as being in
the center, environmental changes are often interpreted as being caused by organizational
action. This egocentric learning may be successful in the short run, but it often occurs at
the expense of strategic success in the long run. In this sense, autopoiesis as a theory of
self-creation of systems also has implications for self-destruction of systems.
For instance, at the beginning of the commercialization process of the Dutch
Railways, managers addressed complaints of passengers by interpreting them from a
technological frame of reference. Becoming more client-oriented was explained in terms
of improving the quality of railroad equipment instead of improving customer-related
services. Because operating trains was seen as their main mission instead of carrying
72
passengers, managers translated the information from a technological viewpoint .
8.2.5 Hidden learning
Learning processes often occur unnoticed (e.g. Brown and Duguid 1991, Ciborra
and Lanzara 1994). This is mainly because the actual processes in which organizational
knowledge is constructed or restructured occurs during non-canonical work practices
(Brown and Duguid 1991). For example, organizational members adjust their work
71
This 'dynamic conservatism' is similar to the concept of single loop learning (Argyris and Schon
1978). Whereas single loop learning occurs within the existing frames of references, double loop learning
questions these guiding principles. The difference with the 'autopoietic perspective' lies within the attributed
origin of this conservatism. Whereas Argyris and Schon argue that defensive routines are the main obstacle,
the autopoietic images consider self-reference as the main barrier to significant changes. Furthermore, single
and double loop learning are constructs that refer to the outcome of learning whereas the autopoietic
perspective on organizational learning directs the attention to the action process of learning.
72
Students attending the course on Organizations and Management, September - december 1995, have
analyzed cognitive change and problems of organizational learning at the Dutch Railways. See for a review
of results of this course (Edel 1996).
141
routines in order to remain aligned with their enacted environments. Because management
is focussed too much on canonical practices, it is often unaware of the actual learning that
is situated in day to day interactions.
The consequence of hidden learning is that the organization engages in audience
learning as discussed in chapter four. Audience learning occurs when the organization
assumes it learns from the actions of the organizational members albeit in practice they
learn from what they assume is happening within the organization. As a consequence, the
link between the actual learning practices, and that which is considered as learning
practices, is severed. As will be argued in section 8.3, this in turn may have consequences
for the organization's ability to innovate. Because actual noncanonical practices have a
practical rather than a formal connection to the world, they are continually developing
new interpretations of that world (Brown and Duguid 1991).
In addition to managers, members of non-canonical communities too are not always
aware of their (re)construction of organizational knowledge (Ciborra and Lanzarra 1994).
Because their personal activities are so much integrated in their day to day context, they
could become blind to changes at the level of the group or organization that their actions
bring about.
8.2.6 Avoiding focussed learning
Avoiding focussed learning requires first of all an understanding of its possible
origins. Next to sheer understanding, focussed learning may also be avoided by improving
the communication between organizational
members
and between
organizations.
Furthermore, avoiding focussed learning calls for an awareness of tendencies that result in
selective search and interpretation.
- Communication
As will be argued in chapter nine, imperfect communication as a result of
specialization and physical conditions can be reduced by improving the communication
between the various units through for example the use of e-mail and Intranet. Job-rotation
is also a way to reduce the occurrence of problematic communication due to job-division.
In general, organizations should try to incorporate redundancy within the organization.
Redundancy assures a certain overlap of knowledge which is believed to be productive for
142
learning (Cohen and Levinthal 1991, Nonaka 1990). According to Nonaka (1990), many
Japanese firms have incorporated redundancy in their actions in order to stimulate
knowledge creation. Redundancy assures more conversations and communication which
contributes to the emergence of a 'common cognitive basis' which enables the transfer of
tacit or implicit knowledge. Because organizational members exchange overlapping
information they are better able to sense what others try to put into words.
Simultaneously, the possible occurrence of self-referential information use, and the
dominance of reference groups as groups that influence the interpretation, calls for more
diversity within the organization. Diversity is needed to interpret information in different
ways and to consider alternative environments as models to imitate.
A mixture between redundancy and diversity may be acquired by adjusting the
design of organizational architecture as well as the ways groups and individuals are linked
to each other. This architecture should acknowledge and even promote a certain level of
autonomy of units. Members of communities should be able to develop their independent
individual beliefs instead of adapting too fast to organizational routines. In other words,
the design should allow for some degree of diversity gained through anarchistic learning.
At the same time, the design should also allow for an interconnectedness through
which the results of actions and beliefs of separate communities are able to spread. This
requires enabling the circulation of stories (Brown and Duguid 1991). Computerized
information systems may support this exchange of narratives, for example through
Intranet.
In themselves, Intranet and Internet provide promising possibilities for avoiding
self-referential use of information while at the same time they enable overlap of
information. Organizations and organizational members externalize their knowledge so that
it becomes public. Because of this externalization, more understanding and knowledge is
gained about the dispersed experiences within the organization. Simultaneously, Intranet
and Internet are information-monitoring systems instead of an information-analysis systems
(March 1994). As a result, using the system may produce surprises and serendipitous
findings which counterbalance self-reference. I will return to this issue in chapter nine.
- Self-awareness
The revelation of hidden learning, cultural conditions, self-referential forces and
the predominance of reference groups, also require self-awareness.
143
Self-awareness creates understanding of existing organizational knowledge and of
the actual learning practices that may (re)construct this knowledge.
Self-awareness
takes place through
73
self-reflection .
In the literature
on
organizational learning, self-reflection is mostly portrayed as a deliberate kind of learning
74
in which psycho-therapy serve as an interesting metaphor (Argyris and Schon 1978,
Swieringa and Wierdsma 1990). As in psycho-therapy, self-reflection involves the process
of becoming aware of ones personal history. In order to 'heal' patients, the therapy is
meant to bring feelings of which the patient is unaware into conscious awareness.
Reflection is the basic technique of psychotherapy in which the therapist helps the client
reflect on his or her emotions to clarify his or her feelings. Likewise, within
organizations, this reflection can take place through the intervention of a 'therapist' or a
'third actor' (van Dongen 1991). Although insightful, this metaphor has some limitations
in that it focusses heavily on curing organizations as the raison d'etre for organizational
learning, whereas here it is argued that there can be many other reasons for organizations
75
to learn .
Again, organizations should not be too self-aware. Sometimes, situations require
foolishness. Trying to be aware all the time of what knowledge exist within and outside
the organization could limit the occurrence of pleasant surprises. Moreover, too much
self-reflection might result in situations in which organizational members interpret
information from collectively agreed upon viewpoints. In their most extreme form, shared
frameworks are important barriers for change and innovation (Raesfeld-Meyer, von et al
1996)
73
Whereas self-reflection is usually connected with 'higher level' learning (Argyris and Schon 1978),
here it is argued that self-reflection is needed during all types of learning, ranging from internal learning to
explorative learning.
74
To be more precise, this concerns humanistic psychotherapy such as client-centered therapies
(Rogers 1951).
75
Self-awareness through self-reflection may result in unlearning (Hedberg 1981). Just as individuals
sometimes need to unlearn negative feelings and behaviors, organizations too may face the need to unlearn
obsolete knowledge. Hedberg (1981) defines unlearning as "a process through which learners discard
knowledge" (p. 18). Unlearning may open the way of new learning to take place. The reasoning is analogous
to Kurt Lewin's (1951) idea that organizational change can best be implemented if a felt need for change is
first created, if an "unfreezing" occurs.
144
Self-reflection technologies - the use of dialogue sessions (Bohm 1990), stakeholder
analysis (Mason and Mittrof 1981), or the use of Group Decision Support Systems (e.g.
Boland et al 1994) - are able to promote self-awareness. But it can also be enhanced with
the use of ethnographic studies (e.g. Roth and Senge 1996). Because of the subtle and
hidden nature of most learning processes, thick descriptions of day to day work practices
done by 'external' researchers could reveal the elusive and unpredictable character of self76
awareness .
The possibility that actors become aware of changes depends heavily upon the
degree of cognitive openness and vulnerability of the actors themselves. It can also depend
on what the system allows, "or, in the words of the poet Keats, on the degree of Negative
Capability they (the actors, MH) are equipped with, that is the capability 'of being in
uncertainties, mysteries, doubts, without any irritable reaching after fact and reason"
(Ciborra and Lanzarra 1994, p 25).
Indeed, fluctuations often go unnoticed because organizational knowledge which
haseproven successful in the past and stimulates actors to unreflectively perform their
routines. Even when novelties or other changes destabilize these frames of references,
actors stick to their old ways, showing limited individual learning skills (Ciborra and
Lanzara 1994). Negative capability is indeed a quality that management and the old-timers
at AZ seemed to lack. The new professional routines brought in by the new group of
system designers was almost ignored instead of being used to question the status quo in
order to change the general organizational routines (Argyris and Schon 1978).
Negative capability can only help to avoid focussed learning when there is some
degree of trust between the various organizational actors.
Mayer, Davis and Schoorman (1995) define trust as:
"the willingness of a party to be vulnerable to actions of another party based on
the expectations that the other will perform a particular action important to the
trustor, irrespective of the ability to monitor or control that party (p. 712).
76
The Xerox Palo Alto Research Center and the Institute for Research on Learning in Palo Alto, CA,
conduct a lot of anthropological research on organizational practices.
145
Consistent with this definition, trust can be considered as the countervailing force of fear
77
that may be the result of the wielding of power (Webber 1993) .
Self awareness requires that members are willing to be vulnerable with colleagues
including subordinates and superiors (Schein 1992). Consequently, some level of trust is
needed in order to stimulate self-reflection which then can prevent focussed-learning.
The idea of trust as being of crucial importance in organizational life is growing in
popularity within management and organization literature (Fukuyama 1995, Handy 1995,
Mayer et al 1995, Peters 1992, Porter 1990, Webber 1993). The notion that a minimal
level of trust is needed during situations of learning in order to increase understanding of
the "theories in use" was first propagated by Argyris and Schón (1978) and was made
more explicit by Argyris (1985, 1990), Dodgson (1993a) and Moingeon and Edmondson
(1996).
Trust is very difficult to achieve. Tom Peters (1992) as one of the many authors
who advocates trust within organizations, devotes a chapter in his book "Liberation
Management" to it:"The Missing 'X-factor'". Peters' explicit attention to the necessity of
trustful relations notwithstanding, he is not able to explain in depth the absence of trust
within organizations and how to promote it. Charles Handy, as another contemporary
trust-champion (1995), believes that an important aspect of trust lies in the size of the
78
group of people working together as well as its more or less enduring character .
Like too much self-awareness, too much trust will not be beneficial. Suspicion for
example may subdue the occurrence of 'group-think' (Janis 1972). Furthermore, groups
that are characterized by mutual trust are likely to become too homogeneous.
In short, avoiding focussed learning requires first of all an understanding of its
origins. Such an understanding in turn facilitates its avoidance. Focussed learning can also
be bypassed by increasing the level of communication through a combination of diversity
77
Trust should not be considered the opposite of power. In fact, by trusting someone, one may allow the
other to use power over the other.
78
This need for trust is becoming increasingly important as professionals and knowledge workers more
and more replace blue collar workers. Because of their professional expertise, these knowledge workers
know much more than there management. Consequently, managing through means of control is becoming
increasingly irrelevant whereas trust becomes the most relevant alternative for organizations to exploit the
knowledge of its members. This need for trust becomes almost a necessity within 'virtual' or network
organizations (Handy 1995, Peters 1992).
146
and redundancy, and increasing the level of self-awareness, combined with negative
capacity and trust within the organization. In the next chapter, I will elaborate on the
possibilities of information systems to avoid the occurance of focussed selection,
interpretation and use of information.
8.3
UNBALANCED LEARNING
Avoiding focussed selection, interpretation and use of information is not the only
prerequisite to stimulate successful learning. As mentioned frequently in part two,
successful learning also requires balancing the various types of learning. In chapters four
to seven, four types of learning were discussed separately. This was done to provide
conceptual clarity. In practice, the four types of learning need to be integrated in order to
improve organizational learning capacity. Table 8.1 outlines various mutual dependencies.
Reading from left to right, the four learning processes represent the process of learning
that is the focus of attention, because for example it is dominant at that time or because it
will be dominant in the future. The columns, reading from top to bottom, represent the
learning processes that contribute to these dominant learning processes. Below I will
further explain the content of all twelve cells.
Using feedback information to advance internal learning (cell 1)
As a result of the closed character of internal learning, unintended conservatism is
a conspicuous phenomenon during this type of learning. Since the organization only learns
from existing experiences, its chances of survival are pretty low if it relies solely on this
type of learning. As mentioned, internal learning is only a conceptually useful construct
since it emphasizes the construction of organizational knowledge within a closed systems.
The organization learns from its own experiences without paying attention to any
environmental reaction that this learning could bring about. Consequently, learning results
in path dependency. The organization continues doing what it always does, ignoring
signals from the environment that may alert the organization to change its current
behavior. The obvious answer to this problem is first of all to allow for feedback from the
environment.
147
B: Learning processes as part of (A)
internal
internal
feedback
learning from
learning
learning
others
X
learning
creative learning
1.
2.
3.
avoiding path
avoiding path
avoiding path
dependency
dependency,
dependency,
promoting
promoting
mutual learning
anarchistic
learning
A:
feedback
Learning
learning
processes
4.
X
5.
6.
interpreting
avoiding path
exploring
feedback
dependency,
feedback
information as
interpreting
information
intended
feedback
information
more thoroughly
learning
7.
8.
X
9.
from
assimilating
avoiding
capitalizing on
others
external
needless
absorptive
knowledge,
institutional
capacity
avoiding too
imitation
much
dependency
creative
10.
11.
12.
learning
creating a
dangerous
dangerous
"prepared mind"
relation
relation,
stimulating "new
combinations"
Table 8.1 Mutual dependencies of the four types of learning
148
X
Using experiences of others to advance internal learning (cell 2)
To avoid path dependency, internal experiences gained through internal learning
may also be complemented by the experiences of other organizations gained through the
learning from others. This integration of internal and external experiences may happen
unnoticed, for example as was the case with the entrance of professionally trained system
designers at AZ. These newcomers had learned their occupational routines through their
education and previous work experiences. In other words, by hiring these newcomers,
external knowledge of other organizations was introduced.
The case story also illustrated the necessity of being aware of the intertwinement of
various learning processes. Since management overlooked this knowledge potentiality, the
information system design section was a victim of unconscious conservatism due to
ignoring experience from others while learning from existing experiences.
This situation highlights the importance of balancing mutual learning. During the
socialization of individuals to the existing organizational routines, people adapt to the or­
ganizational knowledge as they become organizational members. This learning should
however be balanced with processes of learning from others in order for the organization
to learn simultaneously from the new knowledge that individuals may introduce. As March
(1991) argues, organizations are inherently conservative and tend to learn slower than
newcomers. Speeding up the learning process of organizations or slowing down the
learning process of newcomers implies for example a change in training and socialization
practices.
Using creativity to advance internal learning (cell 3)
Path dependency may also be bypassed when creative learning forms part of
internal learning. The surfacing of private experience can for example be complemented
by creative learning processes when one is forced to perceive his or her own experiences
and that of others from different angles. As will be argued in the next chapter,
information technologies such as Groupware may support this process. Creative learning
should also complement internal learning so as to allow for a certain degree of anarchic
learning, as discussed in chapters four and seven. In the course of day to day activities,
individuals may produce new insights that depart considerably from the existing
organizational knowledge (Brown and Duguid 1991). If management overlooks this
process, it cuts itself off from major sources of creativity and innovation.
149
Using past experiences to advance feedback learning (cell 4)
During feedback learning, the organization learns from its own experience by
adapting to environmental demands. Universities, for example, learn to adjust their
curriculum; governmental agencies learn to change their policies. Feedback learning will
likely be more fruitful when it stands on the shoulders of successful internal learning. For
without an awareness of their own experiences, organizations do not know to which
environmental reactions they should adapt and or how to adapt to them. This is needed to
overcome instances of superstitious learning during which the organization learns from the
environmental reactions while these reactions are not dedicated by the organizational
actions but by other exogenous events. Awareness of past experiences may improve the
understanding of the relationship with the environment. Awareness of the organizational
idiosyncratic knowledge is also needed to avoid 'learning under ambiguity'. Furthermore,
this awareness will increase the understanding of the existing patterns and frames of
interpretation.
Using experiences of others to advance feedback learning (cell 5)
Feedback learning can also be liable to unintended conservatism since the selection
of - enacted - environments is often a derivation from the past. The next type of learning
that has been distinguished: learning from others, may to a certain extent limit the
occurrence of this form of unintended conservatism. With learning from others, new
knowledge is diffused and adopted. The organization does not learn from its own
experience but from the experience of other organizations.
Furthermore, integrating experiences of others while adapting to feedback
information is also needed to interpret the information more thoroughly. When for
example, organizations scan their environments through organizational benchmarking, they
become more knowledgeable what 'fellow' or 'rival' organizations are doing or planning
to do. Such an understanding could be of importance to interpret feedback information. To
illustrate this point, a change in customer demands could be interpreted as an organization
specific complaint in which case the organization would likely react by adjusting its
current actions. The same change in customer demands could also be interpreted as a
more structural change of customer demands in general, irrespective of the actions taken
by the organization. Consequently, when organizations are more aware of their
environments, the occurrence of so called 'superstitious' learning that is a conspicuous
150
trap of feedback learning, may be reduced.
Using creativity to advance feedback learning (cell 6)
Path dependency as a result of feedback learning may also be bypassed when
aspects of creative learning are integrated in the process. As such, organizations explore
with the feedback information in order to learn more about their environments and what
demands there exist or will exist. Data-warehousing for example is a technology well
designed to support this creative feedback learning process. Through data warehousing,
organizations are able to explore the data that has been gained from their environment - in
most cases its customers - in order to search for new combinations and possible gaps in
the present supply. Another possibility is to explore various interpretations that can be
given to the same feedback information. In the next chapter, I will discuss the potentials
of Group Decision Support Systems to support this process.
Using past experiences to advance learning from others (cell 7)
In order to enable efficient assimilation of external knowledge with the existing
organizational knowledge, the learning from others should be complemented by successful
outcomes of internal learning. As mentioned in part two, the absence of prior related
knowledge is an important obstacle to successful outcomes of learning from others. As
many innovation writers have argued, successful innovation depends not only on the
knowledge that is diffused from outside the organization but also on the existing
organizational knowledge (e.g. Sahal 1991, Rosenberg and Fristak 1985).
A study of the history of SABRE airline reservation system (Copeland and
McKenny 1988) shows for example how its success has been a result of cumulative
experiences and the difficulties competitors faced in imitating the system. Copeland and
McKenny refer to the notion of "intelligent persistence".
"Intelligent persistence leads to invaluable experience not easily imitated by
rivals. Firms that begin to ride an experience curve ahead of their
competitors realize a head start that will endure as long as new
opportunities continue to be revealed. Technology can always be purchased,
but the same can rarely be said for knowledge" (Copeland and McKenny
1988, p. 368).
151
Another example that highlights the problematic relation between the learning from
others and internal learning is the contemporary trend of organizations to outsource
competencies that do not belong to the organizational core. When business activities are
delegated to external independent organizations,
independent
organizations
organizations learn from
whenever they purchase
these
services or products. Many
organizations for example have decided to contract out their information systems design
and development activities to consultancy firms. This outsourcing can be a dangerous
activity in the event that the organization itself may not have sufficient expertise about
these cut off activities. In other words, a paradoxical situation arises since organizations
mostly outsource activities simply because they do not have the necessary expertise to
perform the activities in house. In such cases however, consultancy firms become rich not
so much because of their expertise in itself but more because they are the one-eyed man in
a country of the blind. Table 8.2 shows this problematic relationship.
Internal expertise
External
expertise
Low
High
Low
disaster
seldom happens
High
likely to succeed
but expensive
successful
Table 8.2 External and internal expertise in relation to outsourcing knowledge
Using feedback information to advance learning from others (cell 8)
When organizations learn from the experience of others without being aware what
effects their own actions may have on their environments, they may become victims of the
power of institutionalized forces. An organization for example may want to become
commercialized because this is a trend within the relevant field (Scott 1983). This
willingness to 'go with the flow' may however be independent from the actual needs
within its environment, such as clients or customers. Forssell (1989), for example,
insightfully illustrated the problems a banking company was facing when it was striving to
commercialize while ignoring the actual needs of its customers.
In a similar vein, the present urge of many organizations to respond quickly to the
152
latest trends within the ever changing world of information technology, often goes beyond
the actual demands of the environment (Davenport 1996).
Using creativity to advance learning from others (cell 9)
Overreliance on experiences of other organizations may also result in unintended
conservatism. The past success of imitating other organizations for example, will likely
enforce future patterns of learning. Past success will negatively influence the probability
of considering alternative models to imitate (Levitt and March 1988).
Balancing this type of learning with aspects of creative learning could diminish this
probability. Through successful outcomes of creative learning, for example by R&D units,
the width of the existing organizational knowledge is extended. When the gains from
creative learning are efficiently diffused within the organization, the absorptive capacity of
the organization to locate and assimilate external knowledge will increase (Cohen and
Levinthal 1990). In such a case
"the investment in irrelevant fundamental knowledge is not directed primarily
towards making discoveries or inventing new policies but to developing the
knowledge base required for profiting from policies and discoveries made by
others" (March 1994p 247).
For example, experimenting with chipcard technology that occurred in the
beginnings of the nineties at LeaseCo had created a certain absorptive capacity which
helped facilitate recognition and understanding of trends within the world of chipcard
technology. Without this past exploratory experience, Johnson would probably not have
been triggered by other actors in the field of business travel and chipcard technology to
connect both worlds and to explore the possibilities of a 'Mobility Pass' as a new service
of LeaseCo.
Using past experiences to advance creative learning (cell 10)
Outcomes of internal learning, or awareness of past organizational experiences, can be
helpful during creative learning. Outcomes of internal learning are also needed to create,
in the words of Pasteur, a prepared mind that favors fortune. Or as Simon (1985) puts it:
"It is the surprise, the departure from the expected, that creates the fruitful
accident; and there are no surprises without expectation, nor expectations
153
without knowledge" (p. 11).
This relationship implies that those who engage in exploration, or experimenting
should be aware of the past organizational experiences.
Using feedback information to advance creative learning (cell 11)
In contrast to all other linkages, the relationship between creative learning on the
one hand and elements of feedback learning on the other, is a dangerous one. During
creative learning it is best to be cut off from feedback information since feedback
information is most often negative in the short run (March 1991).
This implies that at the moment creative learning takes off, organizational members
should be brought in some degree of isolation from day to day activities. Combining this
implication with the previous implication that members engaging in creative learning
should have some knowledge about the past organizational experiences, suggests that
creative learning should not be reserved to a group of people that are totally separated
from the organization.
Using experience from others to advance creative learning (cell 12)
What has been said about balancing creative learning with feedback learning also
applies to creative learning and learning from others. This idea is marked by Polanyi's
comment on one of his contributions to physics:
"I would never have conceived my theory, let alone have made a great effort to
verify it, if I had been more familiar with major developments in physics that were
taking place. Moreover, my initial ignorance of the powerful, false objections that
were raised against my ideas protected those ideas from being nipped in the bud"
(Polanyi 1963, cited by March 1991, p. 85)
Hence, in some cases of learning, ignorance can be a blessing.
But besides this dangerous relation, creative learning may also benefit from
outcomes of imitation processes. Learning from others often stimulate new creative
avenues by combining existing knowledge with new knowledge (Koestler 1964).
Schumpeter for example argued that major innovation are very often a new combination of
existing ideas. As will be discussed in the next chapter, Internet provides interesting
opportunities to create new ideas through combining experience that is made public by
other actors.
154
In sum, treating organizational learning as consisting of only one or two forms of
learning may result in imperfect learning such as path dependency. Efforts to promote
successful outcomes of organizational learning therefore call for a delicate balance
between the various forms of learning.
The question how to balance learning has been deliberately omitted.
This balancing process is definitely not a straightforward endeavor. Standard guidelines
that could serve as early warning signals as to alert the organization that the ongoing
learning process should be balanced by other forms of learning, are not easy to give.
There is no standard rule that tells organizations when to engage in for example more
imitation or more creativity. Judging the issue of balancing depends on an awareness of:
1)
the type of learning that is most dominant at the moment;
2)
the conditions that facilitate and hinders this particular type of learning, as
mentioned in section 8.2;
79
3)
the possibilities of incorporating aspects of other types of learning .
In general however, given that organizations have the inherent tendency to be
conservative, many organizations would improve if they devoted a disproportional amount
of attention to exploration (March 1991).
As mentioned in chapters six and seven, exploration by means of creative learning
and learning from others depends on the availability of slack resources in terms of money,
experience, skills and external contacts.
Money is needed to facilitate experimenting. Besides size and time, economic
conditions which facilitate experimenting can also be subject to institutional constraints or
supports. For example, exploration is facilitated in institutional set-ups where R&D
expenditures can be deduced by means of governmental tax regulations.
Slack resources in terms of experience are needed to increase absorptive capacity
of the organization (Cohen and Levinthal 1990). A diversity in knowledge gained in the
past may facilitate learning from others as well as creative learning.
In addition to diversity in experience, slack resources in terms of a diversity of
79
In appendix I at the end of the thesis, a checklist is presented with general guidelines that may,
amongst others, alert organizations to unbalanced learning processes.
155
skills may also stimulate exploration. Given that foolish behavior may result in the long
run in innovation, recruiting and embracing a small group of people with totally different
backgrounds in terms of experience, culture, education etc. enhances the balancing of
learning.
Finally, slack resources in terms of external contacts enable complementing
learning with the experiences of others. Access to a diversity of informal networks for
example, increases the possibility to use a variety of learning models as sources to imitate.
8.4
CONCLUDING REMARKS
In this chapter, implications are given derived from theoretical arguments put
forward in part two of this thesis. I have deliberately chosen to remain rather general
while addressing their practical significance. As mentioned in the introduction of this
chapter, ready made guidelines may fit one organization but can be inappropriate to
promote successful learning at other organizations.
The implications are mainly useful when an organization wants to assess its current
learning behavior although they may also be of use in case the organization is planning to
learn in the future. Appendix I at the end of this thesis, provides a checklist that can be
used to assess the learning capacity of organizations. This checklist is a summary of the
implications mentioned in this chapter.
The implications can be reduced to four steps that should be taken whenever
organizations want to assess its current or future learning capacity:
1)
Understand underlying dynamics of learning as described in part two. After all, as
we come to understand learning processes better, we can also better assess their
intelligence and efficiency.
2)
Be aware of the possible learning processes that currently occur within the
organization or that will be of relevance in the near future. Since without such an
understanding, it is impossible to manage the process.
3)
156
Avoid as much as possible the occurrence of focussed learning and path
dependency by enhancing diversity of information.
4)
Balance the various types of learning so as to promote successful outcomes of
every type of learning
Improving the learning capacity of organizations, can further be enhanced through
the use of information systems as tools to support the (re)construction of organizational
knowledge. In the next chapter, I will discuss the role of information systems in a learning
environment. This will be done by focussing on the various information intensive
processes that characterize, while analyzing the potentials for supporting these particular
processes with the use of information systems. Attention is also paid to the possible
negative impacts current information systems may have on organizational learning
processes.
157
CHAPTER NINE
INFORMATION SYSTEMS AND ORGANIZATIONAL LEARNING
9.1 INTRODUCTION
Organizational learning is an information and knowledge intensive activity. It is
therefore necessary to discuss as part of its implications how information systems can be
helpful in supporting successful outcomes of organizational learning. Consequently, in this
chapter the role of information systems will be deduced from the theoretical arguments put
forward in previous chapters. I will analyze in what way information systems may enable
or hinder this striving for successful learning. The reader should be warned however that
the chapter lacks any reference to the design of these systems. Although this may very
well be of interest to organizations, I preferred to focus on the social aspects of
information systems rather than the technical aspects.
Information systems are referred to as systems that are able to supply, store, send,
receive, retrieve and process information and that may or may not be supported by
80
information technology . Thus both a computerized decision support system as well as
an age-old library fall under the same heading, although a greater emphasis here will be
on information technological tools.
I will elaborate on six important information intensive phenomena that are of
special relevance to one or two types of learning as discussed in part two, and are related
to other types of learning because of their mutual dependency as discussed in chapter
eight. Table 9.1 shows the relation between the six information processing phenomena and
the four types of learning.
The chapter is built up as follows. First I briefly review the existing literature on
organizational learning and information systems' support and will conclude that there is in
fact a theoretical gap between the two concepts. In order to attempt filling this gap, I will
analyze the role of information systems to support the six information intensive
80
Because, apart from the more functional and technical oriented definitions, there is no general
accepted definition of IS that allows for non-computerized information systems (Blonk van der 1996), I
cannot rely on an already existing definition of information systems.
phenomena.
internal
feedback
learning from
creative
learning
learning
others
learning
externalization of
+
+/-
+/-
+/-
+
+/-
+/-
+/-
+
+/-
+/-
+/-
+/-
+
+
+/-
+/-
+
+
+/-
+/-
+/-
+/-
+
knowledge
objectivation of
knowledge
internalization of
knowledge
information
selection
information
interpretation
idea generation
+
=
of significant importance
47-
=
of importance because the learning should be part of other types of
learning
Table 9.1 Information intensive phenomena and their related type of
learning
9.2
RESEARCH ON ORGANIZATIONAL LEARNING AND INFORMATION
SYSTEMS
A discussion of the implications of organizational learning for information systems
is necessary since the interest within the information systems discipline in organizational
160
81
learning only occasionally surfaces . When information systems researchers make use of
the concept, it is primarily to explain the trial and error nature of information systemsrelated aspects such as information planning (Huysman et al 1994), designing (Hopstaken
and Kranendonk 1990, Salaway 1987) or implementing (Argyris 1977). Although
literature on how information systems may support the very processes of organizational
learning seems to be increasing in importance (e.g. Boland et al 1994, Wijnhoven 1995),
this area of research is still under-represented. A lack of attention from information
systems researchers becomes even more conspicuous when we realize that organizational
learning is an information intensive phenomenon. As is the case for processes such as
bookkeeping and decision making, organizational learning can be described as a process
that involves knowledge acquisition, information distribution, information interpretation
and organizational memory (Huber 1991).
There are at least three reasons why the current information systems discipline falls
short of dealing with organizational learning.
First, conventional information systems theory is still mainly focussed on a narrow
conception of information phenomena, although things are changing with the rise of
technological innovations in communication. In general, information processing is
perceived as a process in which data forms the input whereas information that is
meaningful and useful to the recipient is the output. Computerized information systems are
82
the major tools which provide this information processing process .
All these - and other - phenomena are predominantly approached from an
'engineer' perspective. Because of this perspective, information is mostly conceived of as
quantified processed data. Clearly, organizational learning processes often require the
exchange of information and knowledge that is less structured and quantified.
Furthermore, because the information discipline in general is focussed on a rather
limited conception of organizational processes, it tends to neglect organizational learning
81
There is a vast literature on learning and information systems though this learning concerns machine
learning such as is the case with for example neural networks.
82
Organizations too have been portrayed as information processing systems (Simon 1977, Galbraith
1977). In this case, information is not restricted to quantified data, though the perspectives share with each
other a largely functionalistic tone, thereby treating information as a resource used to reduce uncertainties.
Galbraith (1977) for example sees an organization as a complex system that has to collect and use
information in order to reduce uncertainties regarding their environment. Boundary spanners for example can
be seen as information brokers, capable of gathering information from outside and subsequently diffuse it
within the organization.
161
processes as an information intensive phenomena. Essentially images of a machine and of
an organism are dominant among information systems researchers, leaving alternative
images of organizations almost untouched (Walsham 1991). Although informationintensive organizational phenomena extend beyond managerial processes of monitoring,
control, and bookkeeping, these have been given significantly less information theoretical
attention (Heng and Koh 1992). For example, information plays an important role in
sustaining and changing organizational culture. Instead of quantified data, this information
is mostly processed through story telling, socialization, gossiping, etc. (March and Sevon
1984). Likewise, information plays a significant and indispensable role during
organizational learning.
Although things are changing as a result of the rise of innovative information and
communication technologies such as Computer Supportive Cooperative Work systems
(CSCW), Electronic Conferences, Group Decision Support Systems (GDSS), Internet, and
Intranet, the traditional model of information systems is not very suited to promote
learning. Accountancy, still acting as the dominant model of the information systems
discipline, requires information that is factually correct, reduces uncertainty, and is
delivered to the right person at the right time and the right place.
These requirements only apply to situations of perfect feedback learning. As
mentioned, perfect feedback learning is often an illusion. In practice learning is
complicated as a result of focussed selection and interpretation of information. There is
perhaps an even more important reason why the traditional model of information and
information systems within organizations is not relevant to situations of learning. As
argued in this thesis, there are various ways in which organizations learn, learning from
feedback information is just one possible type of learning.
Another explanation for the neglect of information systems researchers is that the
process of organizational learning has only occasionally been the subject of a thorough
analysis. Without such knowledge it is hard to say something about the use of information
systems to promote organizational learning.
In the remaining sections, the role of information systems during learning is
analyzed by focussing on six important information intensive phenomena that together
characterize organizational learning.
162
9.3
EXTERNALIZATION OF KNOWLEDGE
During the process of externalization, private knowledge is communicated. The
exchange of individual knowledge is traditionally supported by the use of non-information
technological
(IT)
knowledge
systems.
Work-meetings,
committees,
consultative
structures, projects groups and other forms of structured communication may enable the
externalization of private knowledge. Shrivastava (1983) has labelled these systems
'participative learning systems'. Through committees or working groups the organization
is able to pool together the knowledge and expertise of individual members through
communication.
Externalization also occurs during day to day practices within and between
communities of practice (Brown and Duguid 1991, Lave and Wenger 1991). Furthermore,
unexpected
encounters,
coffee-corners,
lunches,
drinks,
all
may
support
the
communication of individual knowledge.
In addition to face to face communication, externalization can be supported by
communication technology such as the telephone, e-mail, and bulletin-boards.
Nowadays the externalization and sharing of individual knowledge is considered as
an important aspect of so called 'knowledge management'. Contributors to the concept of
knowledge management have stressed the importance of 'knowledge management systems'
(e.g. Boersma 1995, Peters 1992, Weggeman and Boekhoff 1995).
Knowledge management systems are designed to advance the sharing of dispersed
personal knowledge in order for management as well as other organizational participants
83
to learn from each other .
Knowledge management systems belong to the family of 'Groupware'. The term
'Groupware' refers to software products that support groups of people engaged in a
84
common task or goal . The software used provides a mechanism for individuals to share
opinions and resources (Turban 1995). Groupware consists of three kinds of user
interaction (Kirkpatrick 1996):
83
McKinsey for example introduced a 'Firm Practice Information System' which reports on lessons
learned by project leaders on particular consulting assignments (Peters 1992).
84
When the focus is not only on the technology but also on the people employing the technology,
Groupware is also referred at as "Computer Supportive Cooperative Work" (CSCW).
163
Communication:
mainly with the use of E-mail and related functions;
Collaboration:
on-line discussion groups and common access to documents and
shared databases;
Coordination:
allowing workers to jointly accomplish specific procedures and
tasks.
Lotus Notes and Intranet are presently two of the most widely used groupware
systems which function as so-called 'knowledge management systems' (Kirkpatrick 1996).
Lotus Notes and Intranet act as group communication environment that allows users'
access and creates shared information (Turban 1995). They provide a workgroup E-mail,
distributed databases, bulletin boards, text editing, document management, workflow
85
capabilities, and access to Internet .
Because their general aim is to provide access to knowledge dispersed within the
organization, knowledge management systems avoid to a certain extent the problem of
dominant coalitions acting as organizational knowledge-gatekeepers, as discussed in
chapter four.
They are also able to objectify knowledge so that personal or locally shared
knowledge is transferred to organizational knowledge. As such, knowledge management
systems function to hold on to knowledge that otherwise would be lost in case participants
leave the organization.
Knowledge management systems also face problems that may limit the
effectiveness of externalization. These problems are also relevant to other information
systems that will be discussed in later sections.
First of all, a lot of knowledge within organizations is of a tacit nature and
consequently difficult to articulate. Since knowledge management systems are not able to
capture knowledge that is of a tacit nature, much of this knowledge will be ignored while
individuals tend to rely too much on the expressed part of the knowledge (Van der Zee
See for the differences between Intranet and Lotus Notes, Fortune July 8, 1996
164
1996).
Next to their focus on explicit knowledge, knowledge management systems
primarily address only a small part of the various types of knowledge within
organizations. Figure 9.1 represents the areas that knowledge management systems may
cover. The dark fields depict the area in which knowledge management systems may be
purposefully designed for the support of the externalization of individual knowledge. The
grey field refers to mostly unplanned and unintended externalization.
Tacit
Explicit
Situated
Embedded
Embodied
Embrained
Enculturated
Figure 9.1 Fields of knowledge covered by Knowledge Management Systems
In particular, knowledge management systems only support the exchange of
embedded, explicit knowledge that can be transferred into encoded knowledge or
information.
Embedded knowledge is knowledge that resides in systemic routines such as rules,
86
technologies, and procedures (Blacker 1995) . Just as tacit knowledge stands in contrast
to explicit knowledge, embedded knowledge can be considered the opposite of 'situated
86
Organizational learning processes that involve the learning of embedded knowledge are for example
referred to by Levitt and March's (1988) development of the notion of organizational routines. Organizations
that are predominantly focused on this type of knowledge are what Blacker (1995) call "Knowledgeroutinized organizations", with low skill requirements and typically capital, technology, or labour intensive.
These organizations can be labelled "Machine bureaucracy" to use the typology of Mintzberg (1983) to
typify such traditional organizations.
165
9
knowledge . Situated knowledge is located in the practice and interaction of individuals
(Pentland 1992). By dividing knowledge into embedded versus situated knowledge, it is
recognized that knowledge does not always reside somewhere, for example in heads or
information systems; knowledge may well be created during interaction.
9
9
Furthermore, this embedded knowledge can both be 'embrained or 'embodied .
Embrained knowledge is 'knowledge about' (James 1950) and depends on cognitive
abilities, while embodied knowledge is 'knowledge how' (Ryles 1949) and is action87
oriented, such as skills (Blacker 1995) . Both embodied and embrained knowledge are
mostly only partly explicit.
88
Encoded knowledge is information conveyed by signs and symbols . Forms of
encoded knowledge may be both written and generated as electronic information and is
always explicit.
To a lesser extent, knowledge management systems may also channel enculturated
89
knowledge . Enculturated knowledge refers to the shared understandings, is mostly of a
tacit nature, and concerns things such as language, symbols, rituals, norms and values.
Externalization of enculturated knowledge occurs through the very structure of the system.
For example, information systems influences what information and knowledge is
considered as relevant and what will be considered as irrelevant. I will return to the
impact of information systems on culture and vis-versa when addressing the process of
internalization.
87
Organizations that emphasize embrained knowledge are typically "Knowledge Intensive Firms"
(Starbuck 1992) such as software consultancy. According to Blacker (1995), Argyris and Schon (1978) as
well as Senge (1992) refer to organizational learning processes that predominantly involve the learning of
embrained knowledge.
Organizations that emphasize embodied knowledge are expert-dependent organizations or
'professional bureaucracies' (Mintzberg 1983) such as hospitals. Organizational learning processes that
involve the learning of embodied knowledge are for example referred to by Hirschhorn (1984) who
illustrated that operators' tacit understandings of machine systems are more important than their general
knowledge.
88
We could think of information intensive organizations such as administration offices as organizations
typically focused on encoded knowledge. Organizational learning processes that predominantly focus on the
learning of encoded knowledge are for example referred to by Zuboff (1988) who analyzed the informating
power of IT.
89
Organizations that emphasize explicit enculturated knowledge are often communication-intensive such
as McKinzey & Company (Peters 1992). Organizational learning processes that involve the learning of
enculturated knowledge are for example referred to by Senge's discussion on the importance of a shared
vision.
166
Table 9.2 provides examples of each type of embedded knowledge in a tacit and in
an explicit form. Because of the emergent nature of situated knowledge, it is not possible
to provide examples of knowledge that is situated. Consequently, the examples given in
table 9.2 all concern cases of embedded knowledge.
explicit knowledge
tacit knowledge
embrained
facts: 1 + 1 = 2
repressed experiences
embodied
manuals
riding a bicycle
enculturated
corporate vision
implicit norms and values
Table 9.2 Examples of various types of knowledge
Thus, by overreliance on knowledge management systems as the unique source of
knowledge within the organization, one runs the risk of ignoring the importance of
90
situated knowledge, tacit knowledge and forms of enculturated knowledge .
A final reason which makes knowledge management with the support of
computerized systems problematic, is that not everyone is willing to share his or her
knowledge with others. As a result, the stored knowledge base may not represent the
actual knowledge that is present within the organization. This is further complicated by
the fact that the updating of knowledge management systems also requires time and effort
by organizational participants, something which heavily depends on its perceived returns.
90
Another possible problematic aspect of information systems such as Intranet or Lotus Notes is that it
implies that no face to face communication is needed in order to exchange information. According to the
information richness theory (Daft and Lengel 1986), the exchange of information without face to face
communication can be problematic, especially in case information systems support the exchange of rich
information. Information richness is defined as the ability of information to change understanding within a
time interval. In order of descending richness, the media classifications proposed by Daft and Lengel (1986)
are (1) face to face, (2) telephone (3) personal documents such as letters and or memos, (4) impersonal
written documents and (5) numeric documents. However, these traditional communication media have the
potential to be supplemented with or replaced by new electronic communication media such as e-mail, voice
mail, video conferencing, electronic bulletin boards, which compliates the validity of the theory. Email for
example can also support rich information (Lee 1994). Consequently, more research is needed on the
possible consequences of externalizing knowledge without face to face communication.
167
In sum, although systems that are believed to support so called 'knowledge
management' are capable of supporting the management of externalizing knowledge, they
are focussed on the externalization of explicit embedded knowledge and assume people are
willing to make their private knowledge public. Hence, organizations should not rely
solely on these computerized systems when they want to learn from their members. Other
non-IT-based systems such as meetings, stories, communities of practice, etc. are
probably more important information systems that support the externalization of private
knowledge.
9.4
OBJECTIVATION OF KNOWLEDGE
When knowledge has become externalized, it will be objectified into collective
knowledge (Berger and Luckman 1966). In terms of organizational learning, this process
takes place when knowledge is stored in the organizational memory. Knowledge must
have certain characteristics for it to become organizational knowledge and to be retained
by the organization. Duncan and Weiss (1979) argue that organizational knowledge must
be communicable or understood by others; it must also be 'consensual', or accepted by
others for its validity and utility. I would add that these 'others' are usually members of
dominant coalitions which act as organizational knowledge-keepers. These dominant
coalitions may be very influential when deciding about the content of the organizational
memory.
A rather broad conception of organizational memory is used here. Although
organizational knowledge is always embedded, it may range from explicit, embodied and
embrained knowledge such as manuals and organizational annuals, to tacit enculturated
knowledge such as language, rituals, symbols. Information systems, such as libraries,
Management Information Systems (MIS), and databases are pre-eminently suitable to
function as repositories of explicit organizational memory. However, what has been said
above regarding the limited possibilities of transferring private knowledge into encoded
knowledge, also applies to the possibilities of computerized information systems to
function as organizational memory.
Although the literature that addresses organizational memory information systems
(OMIS) is growing (see for a review Stein 1995, and Walsh and Ulson 1991), it is still in
168
91
its infancy .
In general, existing literature on OMIS tends to neglect more (psycho-)sociological
aspects of phenomena. For example, OMIS are prone to power issues. The question who
determines what knowledge should be considered organizational knowledge and as such
should be stored in the OMIS, is one of importance but has not been addressed fully. The
same is true for the subjectivity of organizational memories. OMIS's provide
interpretations of history rather than an objective collection of information from the past.
The literature on OMIS also approaches organizational memory as a rather static
outcome of learning processes. Researchers tend to overlook the fact that organizational
memory is always reconstructed the moment the embedded knowledge is used in practice.
In contrast to computerized information systems that tend to stabilize these past
experiences (Hedberg and Johnson 1978), non-IT based OMIS change constantly. The
memory of organizations is for a large part captured in fuzzy systems such as stories. And
because of its fuzziness, the content of the memory develops and changes over time.
Stories for example, are told and retold in organization. As Sims (1996) concludes:
"If we want to look at organizational learning, this change in the stories that are
told (...) will be a particularly fruitful place to look at both for hearing and understanding
both the content and the process of organizational learning" (1996, p. 6).
Clearly, more research is needed on the role and impact of information systems to
objectify organizational knowledge. What has been said for externalization of knowledge,
also applies to objectivation of knowledge: the organizational memory is too much a fuzzy
phenomenon to be supported solely by explicit information systems. Experts, informal
networks, stories and other non IT based information systems are probably more
important in determining the content of the organizational memory.
91
Stein and Zwass (1995) define an organizational memory information system (OMIS) as "a system that
functions to provide a means by which knowledge from the past is brought to bear on present activities" This
definition is followed by the following sub-sentence: "thus resulting in increased levels of effectiveness for
the organization". This criterium of effectiveness is based on a functionalistic model derived from the four
effectiveness functions identified by Parsons (1959): integrative function, adaptive function, goal attainment
function, and pattern maintenance function. I do not agree with this latter part of the definition:
organizational memory may very well result in a decrease of effectiveness, for example when the knowledge
becomes obsolete or when organizations rely too much on their memory.
169
9.5
INTERNALIZATION OF KNOWLEDGE
Objectified knowledge will in turn be internalized by organizational members at the
moment it is used in practice. Internalization of knowledge takes place through the
learning of history and the learning by imitating colleagues. Learning from history often
happens through story-telling, gossiping, and idle talk (March and Sevon 1984). History
can also be transformed into explicit information systems in the form of manuals, for
example, which can be used for training purposes. Learning by imitation can be supported
by cooperation and apprenticeship (Lave and Wenger 1991). Again, the role of
computerized information systems is limited during internalizing explicit knowledge.
(Computerized) information systems do however play an important role in the
internalization of tacit enculturated knowledge. Cultural aspects can always be found in
information systems since information systems are ultimately a representation of reality
and therefore also of the culture (Tibosch and Heng 1994). Information systems provide
means of representing reality through a set of concepts and symbols, and in so doing,
information systems can be considered as a medium for the construction of social reality
(Orlikowsky and Robey 1991). Based on Giddens' Structuration theory, Orlikowsky and
Robey (1991) argue that IT makes it possible to institutionalize the interpretation
framework. While using the three modalities offered by Giddens: interpretive schemes,
resources, and norms they demonstrate that by using information systems, "users draw on
embedded knowledge, assumptions, and rules and through such use reaffirm the
organization's structure of signification". By using information systems, users also "work
within the rules and capabilities built into them, and through such use reinforce the
organization's structure of domination". The use of information systems also assures that
"users work within the authorized options, values, and sanctions built into them, and
through such use sustain the organization' structure of legitimation" (Orlikowsky and
Robey 1991, p. 161).
An illustration of this process of internalization through information systems is
offered by Walsham (1991) while referring to the implicit function of accounting systems.
Accounting systems are predominantly used to set targets, to monitor performances and to
identify and correct failures. However, these accounting systems are only one way of
looking at the world which institutionalizes organizational boundaries and emphasizes
170
certain numerical data. As such they can be seen as "institutionalizing the dominance of
financial information" (Walsham 1991, p. 92)
Hence, information systems are clearly more than a technical construct. Everyone
who is confronted with information systems in one way or another should recognize this
in order to avoid or at least to be aware of possible (unintended) consequences such as
domination and manipulation (Tibosch and Heng 1994).
9.6
INFORMATION SELECTION
Selecting information from the environment is a critical stage during learning, and
in specific during feedback learning and learning from others. During feedback learning,
information is collected from environmental actors, such as clients, customers, suppliers,
and other stakeholders. In case of learning from others, the environment has a much
broader scope and includes actors within the ecology of organizations. Enactment
decisions such as what environments are relevant and what information is useful from
these environments, are important to these two types of learning.
The literature on information systems and feedback learning is extensive, although
seldom expressed in the same words. Management information systems (MIS) are pre­
eminently suited to gather feedback information. Accountancy forms the prevalent model
of these MIS's. MIS's as accounting systems provide 'information' on the past
performances of the organization, current operating conditions and future projections. One
92
of their main functions is to detect and correct errors (Argyris and Schôn 1978) .
Inter-organizational
information
systems, while reducing the organizational
transaction costs, may also yield feedback information and information from others such
as is the case for example with Computerized Reservation Systems in the airline industry
(Christiaansen 1994).
Executive information systems (EIS) also support the process of collecting
92
In fact, when problematizing the role of IS in terms of organizational learning, Argyris (1977) refers
to the defensive theories in use when designing and implementing information systems. He however does not
problematize the very structure of IS's: IS's are conceived of as traditional accounting systems, designed to
identify failures and to correct them.
171
information from the environment during both feedback learning as well as processes of
learning from others. It is a computer-based system that serves the information needs of
top executives. It helps management to scan both the internal as well as the external
environments. While internal information is generated from various business units,
external information comes from sources such as on-line databases, newspapers, industry
newsletters, governmental reports, personal contacts and so on. These environments are
scanned by the executives themselves, by staff, and/or by machines. EIS's are different
from decision support systems and MIS's in that it is not primarily meant to support
decision making. Rather, the aim of a EIS is to provide as much information about the
environment as possible (Turbain 1995). In other words, EIS's are pre-eminently suited to
support learning from others. The choice as to which environments are considered
relevant and which are not, is however already an important decision in itself and can be
subject to self-referential information use, as will be discussed below. In the words of
March (1994) through such information systems, an organization learns from an ego of
itself.
Information systems that support organizational imitation are systems that have the
latent function of 'disease carriers'. Management journals and books, conferences,
Business school courses, personal networks, consultancy firms, are examples of non-IT
based systems that enable the diffusion of external knowledge. With the use of these
systems, organizations learn to gain or maintain their legitimacy in their organizational
field.
In the previous chapter, I argued that one of the significant problems organizations
face when learning, is that this process is characterized by focussed selection of
information. This focussed selection of information is a conspicuous tendency of
computerized information systems. Miller for example refers to so called "focussed
information
systems"
that
"institutionalize and routinize
gaps in
organizational
intelligence".
"Management information systems do not track the things managers believe
to be unimportant or unchanging, but instead focus attention on what is
thought to have mattered in the past. (..) And in many successful businesses,
executives develop the self-assurance to home in very precisely on what they
believe explains their success. Their information systems then fix upon this
172
and ignore everything else. (Miller 1994, p. 330)
Besides history, this focussed selection may also be a result of the dominance of
reference groups who are able to influence enactment processes. What the group believes
to be an important environment from which feedback information should be gathered or
which should be imitated, may determine the actual learning process. In addition,
specialization of the task structure will buffer the external knowledge diffusion and
influence the exposure to information. Of special importance is the dominance of self
referential forces in influencing the selection of environmental information.
The self-referential image of learning makes 'narcissistic' use of information
conspicuous. The selection of information is guided by existing goals, values and
opinions. The "searched for" information shapes the identity and as such influences the
ego-centric, conservative learning processes of organizations. Executive Information
Systems for example are designed to reproduce the organizational identity by directing
attention towards pre-determined elements in the environment. Elements that do not
contribute to a (re)production of this identity will likely not be accounted for. Vital
information which is alien to the frame of reference of an organization is filtered out
before it reaches consciousness, or is reinterpreted or 'rationalized' so as to remove
discrepancies.
Walsham (1991) provides an example from the Vietnam War taken from
Halberstam (1972) to illustrate this self-referential functioning of information systems.
Halberstam describes how the internal organization of the American intelligence gathering
operation in Vietnam was inadequately structured to cope with an understanding of the
evolving conflict in Vietnam.
"Senior decision makers in the American military and political hierarchy
concentrated on the 'information' coming from the field rather than
questioning, until it was too late, the adequacy of the information systems
structure which was supplying their own self-referential view of the world"
(Halberstam 1972, p. 91).
All these forces result in focussed learning and influence the design of information
systems. And because organizations use information systems to view the environment,
they influence the process of organizational learning. Furthermore, once (computerized)
173
information systems are designed and implemented, they tend to freeze these pictures of
the environment (Boogaard 1994).
Hedberg and Johnson (1978) observe that information systems incline to thwart
organizational scrunity and filter away relevant uncertainty, diversity, and change
indicators. They argue that current information systems do more to stabilize organizations
than to destabilize them. All information systems contain an implicit model of the world
93
which may become outdated .
An implication for the development of information systems is the need for a
constant reflection on the adequacy of the structure of those systems (Walsham 1991).
Although this implication sounds trivial, in practice it will be difficult to achieve, certainly
when we consider the problems of the inflexibility of information systems and the coevolving 'soft ware crisis" (Boogaard 1994). Clearly, self-reflection to avoid self-reference
cannot be built into software; self-reflection should come from its users and designers.
Another option to avoid focessed learning as a result of information systems use is
to balance the use of these decision support systems with information systems that are
designed to stimulate surprises and serendipity. To put it differently, in order to avoid
focessed learning as a result of information systems use, organizations should balance
feedback learning and learning from others with aspects of creative learning. I will return
to this issue of information systems supporting idea generation in section 9.8.
9.7
INFORMATION INTERPRETATION
In addition to collection of information, the interpretation of information is also of
due importance during learning.
Designers of information systems may already influence the way information will
be interpreted. The same information can be interpreted differently when it is presented
for example in quantitative or qualitative form, by means of graphs or by means of
stories, through verbal modes or written modes. These degrees of information richness
93
Hedberg and Johnson (1978) have addressed the problem of inflexibility of information systems by
advocating the design of 'semi confusing systems'. In the remainder of their article, the Hedberg and
Johnson (1978) propose ambiguous alternatives to destabilize organizations and IS among which are built-in
'early warning systems'.
174
influences the learning process (Daft and Lengel 1986). The rise of multimedia
technologies suggests a possible reduction of focussed interpretation caused by limited
information display (Boland et al 1994).
The issue of interpreting information has received more attention when approached
from the level of semantics than from the level of syntactics (Boland 1987, 1991, Stamper
1992). Semantics deals with meanings and signification and belongs to the realm of
hermeneutics.
Hermeneutics is the study of interpretation, especially in the process of coming to
understand a text (Boland 1991). When hermeneutics is used to study information systems,
the output from the system would then be viewed as a text being read and interpreted by
the information systems user.
"The output of an information system is an unfamiliar text to be read, interpreted
and made meaningful by those who use it in ways that will always surpass any
clear representation its system's creators had in mind" (Boland 1991, 440).
A hermeneutic perspective on information systems provides insight into the
problems of focessed learning, and learning under ambiguity in particularly. Again,
reference groups, self-referential forces, and specialization significantly influence the way
information is interpreted. Information interpretation is never an objective activity.
Individuals create interpretations, for example, according to the (cultural) norms that are
prevalent within their organization or group. Significant others for instance, can be
influential in shaping the interpretations of others (Smircich and Morgan 1992). Selfreferential forces too influence not only what is considered as important but also how one
should make sense of the information.
As a result of these tendencies, users of information systems may perceive the
same output from different perspectives. The existence of various interpretations has long
been considered as a problematic situation that should be avoided as much as possible.
Writings on the 'corporate culture' and 'shared vision' for example, advocate the
implementation of one single interpretation frame to be used as a meaning provider within
the organization. As mentioned in previous chapters however, the existence of one
dominant way of looking at the world has serious pitfalls. Organizational learning calls for
diversity, not only in terms of a heterogeneous staff but also in terms of multiple
175
viewpoints. In terms of the interpretation of environmental information, this process
requires the exchange and appreciation of each other's perspectives. Through this
exchange, alternative interpretations can emerge which may avoid focessed learning.
Group decision support systems are well equipped to support this process of reflecting on
various interpretations.
Boland and his colleagues (Boland et al 1994, Tenkasi and Boland 1996) for
example, propose the design of IT to support distributed cognition.
"Distributed cognition is the process whereby individuals who act autonomously
within a decision domain make interpretations of their situation and exchange them
with others with whom they have interdependencies so that each may act with an
understanding of their own situation and that of others" (Boland et al 1994, p.
457).
Applications of IT then assist individuals in making interpretations of their
situations, reflecting on them, and engaging in dialogue about them with others.
An example of such a system is "Spider" (Boland et al 1994), a software
environment for distributed cognition. With the use of Spider, actors store their own
interpretation of a particular situation into a knowledge base. The interpretation can be
represented by spreadsheets, cognitive maps, notes, dialogue boxes, and graphs. Each
actor then exchanges his or her uniquely respresented interpretations.
Spider is an example of GDSS and is helpful in exchanging different definitions of
the same situation. As will be mentioned in the next section, GDSS are also useful in
generating new ideas.
9.8
IDEA GENERATION
The final information intensive phenomenon that will be discussed is that of idea
generation and is mainly suited to support creative learning. Nevertheless, in line with the
arguments put forward in chapter eight, other types of learning should incorporate aspects
of creativity in order to become more successful. This also means that in order to avoid
focessed information interpretation and selection, organizations should balance their use of
information systems that support feedback learning and learning from others with
information systems that are well-equipped to foster creative learning. To put it
176
differently, information systems that promote idea generation discussed in this section,
also may promote successful outcomes of feedback learning and learning from others.
Simulation techniques such as scenario planning have often been considered as
important (management) tools to stimulate so called "generative learning" (e.g. De Geus
1988, Isaacs and Senge 1992, Senge 1992, Stata 1989, Vinnix 1990). In fact, the
'Organizational Learning Centre' at MIT, Boston, has developed special 'learning labs' in
which top management of various companies are able to make use of simulation software
to explore possible future avenues (Senge 1992). Simulation techniques are focessed on
learning about multiple enacted futures by trying to tease out the future events which they
realize through posing 'what if questions. However, this exploration is based on an
already formed perspective and predetermined parameters. Hence, during simulation
techniques such as scenario planning one defines ex ante from what to learn. In other
words, these systems stimulate exploring within predefined solutions which limits the open
character of idea generation.
A more unrestricted form of idea generation is open brainstorming and exchange of
ideas. This can be supported by Group Decision Support Systems (GDSS) which again
belongs to the family of Groupware. GDSS is an interactive computer based system which
facilitates the solution of non-quantitative, unstructured problems, and facilitates electronic
brainstroming (e.g. Gallupe and DeSantis 1988, Nagasundaram and Bostrom 1994,
Nunamaker et al 1987). Through electronic brainstorming, ideas are exchanged in order
for new ideas to emerge. By building on each other's ideas, individuals get creative
insights they did not have before (Turban 1995).
Besides the problems mentioned when dealing with knowledge management
systems, a GDSS has also its own problems.
First, behind a GDSS lies a rather harmonious perception of social phenomena,
since it takes the view that people have enough empathy to consider and appreciate each
others standpoint.
Secondly, although a GDSS is often believed to have built-in mechanisms that
discourage the development of destructive conflict, miscommunication, or "groupthink'
(DeSanctis and Gallupe 1987), it remains to be seen to what extent IT can seriously
diminish the occurrence of negative group behaviour. As was illustrated with the AZ story
177
and theoretically argued in chapter eight, miscommunication is often a result of social
psychological forces that are deeply ingrained in the thinking and acting patterns of
organizational members. For example, already during the process of socialization, people
tend to adopt the beliefs of their personal reference group. The introduction of information
systems such as GDSS's will only slightly filter away these influences.
Designers of GDSS's as well as its users should be aware of various - often hidden
- factors that may complicate organizational learning rather than promote it. The literature
on GDSS's which supports the exchange of various interpretations is rather optimistic
about the possibilities of IT to encourage mutual understanding. The area needs more
indepth qualitative research of the use of these systems in real practice. Then, the actual
possibilities of GDSS's during learning might come to the surface.
Up until this point, the focus was mainly on decision support systems. Although
these systems are capable of deriving alternative solutions without any constraints such as
is the case with simulation techniques, they are not well suited to explore alternative
problems. Exploring alternative problems requires different information systems than the
systems discussed so far.
Whereas the rationale behind decision support systems is reducing uncertainty, the
rationale behind these alternative systems is to reduce certainty by extending the scope of
information that might possibly be relevant in an organization's future. Hence these
alternative information systems consist of information that has not been thought of until it
is gained. This implies that one cannot specify the information requirements ex ante.
Although these ideas on the information requirements of learning fully contradict
the traditional assumptions of mainstream information systems literature, they are not
totally absurd. In fact, libraries, as one of the oldest and most universal information
systems are based on these very information requirements (Heng and Koh 1992). Libraries
consist of information sources whose relevance cannot be determined beforehand. In fact,
the greater part of a library's inventory will never even be used at all. Librarians, in
contrast to information managers, do not determine in detail the information requirements
of potential users before they conclude whether an information source has relevance and
thus should be incorporated or not. Therefore, libraries serve the purpose of providing a
large reservoir of new knowledge most of which is irrelevant at the present but could be
of relevance in the future.
178
Another important characteristic of libraries is that their use often leads to unex­
pected or serendipitous findings (Foskett 1984). Although a lot depends on the way the
books are arranged as well as the pleasure the individual has in visiting a library, browsing along the shelves and using cross-references - provides knowledge one never had
thought of or which has been forgotten.
"Those who confine their interpretation of information to its narrowest sense of
factual data seem to forget that browsing among the shelves of a good library
provides a conspectus of any field of knowledge far wider than the compass of one
individual mind, and offers a choice of approach and treatment which can lead to
what W.I.B. Beveridge calls a 'eureka situation'(Foskett 1984, p. 53)."
The image of a library as an information system serves as an interesting model and
provides us with two central features of information systems that promote idea generation.
First of all, information systems that stimulate idea generation should contain an
inventory of knowledge without the constraining issue of relevance.
Secondly, these information systems should provide the opportunity to encounter
unexpected and serendipitous findings. This means that the supply of information should
94
not be arranged too much so that it might become focessed information systems .
The rise of Internet and Intranet facilities provide excellent possibilities for the
support of unfocussed or explorative learning. In the case of Internet, the supply of infor­
mation is world wide. In case of Intranet, the supply of information is basically company
or business-wide although most Intranets provide access to predetermined information
sources that are offered by Internet. Like libraries:
1
Internet and Intranet consist of a large reservoir of knowledge.
2
Internet and Intranet offer the opportunity to make cross-references and to come
across unexpected or previously unknown sources of knowledge.
3
Existing personal knowledge serves as an important platform from which
94
Besides the issue of irrelevance and serendipity, libraries contrasts traditional information systems such
as MIS's in that the information in a library can be factually wrong, and the knowledge and opinions are
often contradictory.
179
exploration takes place.
4
Enthusiasm of the user is important in determining whether he or she will make
full use of the opportunities the system offers.
IS that support these requirements could be called Alien Information Systems (AIS)
(Huysman et al 1994). The salient characteristic of these AIS's is that they capture
information whose relevance cannot be determined beforehand and that they are able to
stimulate chance and serendipity. Hence, libraries, Internet and Intranet facilities offer
excellent examples of the role of AIS's as means to avoid focessed information
95
selection .
Besides the issues of irrelevance and serendipity, information systems that support
idea generation such as AIS's should also enable a different kind of communication
process than the communication during other learning processes. Communication during
idea generation is not only needed to facilitate the flow of information from one person to
the other. Communication during idea generation also assists the learning with each other
and the connection of uncoupled fields of knowledge. In the latter case, communication
serves more as a sort of platform from which new ideas emerge. This was illustrated by
the case story of LeaseCo. The communication between Johnson and his network partners
was more one of brainstorming than one of trading knowledge. In addition, while
communicating his dream of a Mobility Pass to almost every one he came across, he
hoped that his knowledge would flow to places he could not think of beforehand. In other
words, by communicating his idea forcefully, uncoupled fields of knowledge were
connected.
Hence, informal networks are a good example of information systems that promote
this kind of communication. As the Lease Co story illustrated, informal networks seems to
be an important information system to search and gain new knowledge. Certainly,
95
Internet and Intranet do however have their own problems that may limit free access to information.
Imagine the situation in which every one is using the internet to imitate each other. In such an extreme case,
it is not hard to imagine a scenario in which one will think twice before making the outcome of this learning
process public through Internet. Free access to information through the use of the Web is an issue that is
currently one of increasing concern. Its consequences for learning and its related dilemma between creative
learning and learning from others, certainly deserve more research attention.
180
informal networks resemble to a great extent libraries, Intranet and Internet in their ability
to search and create new knowledge, and can therefore be ranged on the side of AIS's.
Clearly, more research is needed on the possibilities of AIS's and in specific
Intranet and Internet to support (creative) learning.
9.9
CONCLUDING REMARKS
In this chapter, I have tried to analyze in what ways information systems can
promote successful learning. This is done by perceiving learning as consisting of various
information intensive processes. These six processes in turn served to explore the role of
information systems.
Information systems provide practical tools for organizations and managers to
promote organizational learning. Amongst other things, their possibilities are related to the
degree in which these systems depend on information technology, to the degree in which it
stimulates focessed learning, and to the degree in which the organization balances the
various types of learning and their related information systems.
Although organizational learning is an information intensive phenomenon, its
relation to information systems has only sporadically enjoyed research attention. As a
result, this present attempt was fairly explorative which has consequences for its scientific
quality. Much more research is needed on the possibilities of computerized information
systems such as GDSS, Intranet, Internet to support organizational learning. But next to
computerized information systems, organizations rely heavily on information systems that
are not necessarily supported by information technology. These systems, such as
Organizational Memory Information Systems, Knowledge Management Systems, and
Alien Information Systems, also need more research attention.
To be sure, it was not my intention here to provide implications for information
systems design. I would like to leave these technical aspects to future researchers.
181
CHAPTER TEN
CONCLUDING REMARKS
10.1
INTRODUCTION
This thesis began by reviewing existing literature in search of a theory of
organizational learning. This effort yielded many valuable though diverse ideas of the
concept. It also showed that contemporary theories on learning are subject to one or more
biases. In an attempt to create an alternative perspective that integrates the existing
valuable aspects of the literature, challenges the various biases, corresponds with
empirical findings, and exploits relevant ideas put forward within other theoretical fields,
a typology of four mutually dependent learning processes was introduced. Subsequently, I
discussed the implications of this alternative perspective on organizational learning for
organizational practitioners as well as for the discipline of information systems.
In this final chapter, some concluding remarks are given. The chapter starts with a
summary of the basic arguments of this thesis. This will be followed by discussing to what
extent this work can be considered a post-modern approach to organizational learning.
Subsequently, the question will be addressed in what way this effort contributes to existing
knowledge on organizational learning. The chapter ends with several recommendations for
further research.
10.2
SUMMARIZING THE ARGUMENTS SO FAR
Organizational learning is a concept that has become more and more popular
within both corporate and academic worlds. Although many articles and books are
dedicated to the subject, there is still a lot of confusion regarding the learning of
organizations. Consequently, what is needed first of all is to try to integrate the various
ideas surrounding learning. This is especially important since every single perspective on
learning sheds light on different yet valuable aspects of the concept. Chapter two
identified six perspectives on learning.
The adaptation perspective is one of the oldest perspectives and considers learning
as a process of adapting to environmental demands. The incremental innovation
perspective understands learning as the diffusion of external knowledge; its success is
influenced by the organization's past history. The assumption sharing perspective is
mainly cognitively oriented and assumes that learning takes place when individuals share
their private beliefs and question organizational frames of references. The organizational
knowledge perspective perceives learning as an information processing phenomenon
during which organizational knowledge is developed. The learning organization
perspective focusses on a specific organizational form that is able to promote successful
outcomes of learning. The social constructivist perspective emphasizes informal learning
processes that take place during day to day activities within communities of practice.
Every perspective has its valuable point. Nevertheless, purely combining these
valuable ideas would not generate a suitable alternative perspective since all six
perspectives have a tendency to be biased in certain directions and at the same time, they
may overlook alternatives. Chapter three identified five biases. By challenging these biases
while making use of the valuable existing ideas on learning, the attempt was made to
introduce a more integrative perspective on learning.
The 'improvement bias' refers to a tendency to assume that learning results in
improvement, intelligence or innovation. This improvement bias is a result of an outcome
perspective on learning. In order to overcome this bias, learning was approached in this
thesis from a process perspective. A process perspective reveals the dynamics of learning
and the potential of learning defects. As a result, learning may also end up in inertia or
even destruction.
The 'individual learning bias' refers to a tendency to focus on individuals as
learning agents. In order to overcome the individual action bias, the use of the
institutionalization theory was proposed where action and structure find each other in a
reciprocal relationship.
The 'systems thinking bias' refers to the predominance of systems thinking as the
model to explain the motives for learning. Organizations learn in order to stay aligned
with their environment. Hence, aspects other than the common 'reaction to environmental
responses' are not taken into account. Plain chance events, exploration, the will of
managers to 'actualize' themselves, the wish to take risks etc. are all important reasons
why organizations learn.
The 'planned and strategic learning bias' refers to the assumption that learning can
184
be planned for and can even be used for strategic purposes. It is assumed that
organizations know what they want to learn tomorrow, that they can anticipate these
learning needs even if this involves radical changes. What these writers tend to overlook
is that learning often occurs unnoticed and accidentally and that because of organizational
history, radical changes are difficult to make. In this thesis, learning was considered an
evolutionary process made up both of stochastic as well as purposefully planned events.
The 'one/two-sided learning bias' is present within almost all literature on learning
and refers to the tendency to focus on only one or two aspects of learning. In case of the
latter, these two aspects are seen as discontinuous while one is inferior to the other. Single
loop for instance, is inferior to double loop, adaptive learning is inferior to generative
learning. In order to provide a broader and integrative perspective than is the case with
the literature on learning, four mutually dependent types of learning have been introduced.
The six perspectives described in literature on learning have made possible the
identification of these four types of learning.
The alternative perspective introduced in this theses, assumes that the process of
organizational learning is an evolutionary process whose outcome depends heavily on the
existence of learning imperfections. The process of learning can be of a different nature,
depending on the sources of (re)constructed organizational knowledge. The types of
learning that have been identified in part two of the thesis are: internal learning, feedback
learning, learning from others and creative learning.
Internal learning is the basic process of learning during which the organization
learns from its members and the members from the organization. Feedback learning
occurs when the organization learns from feedback information taken from the
environment. Learning from others occurs when the organization learns from the
experience of other organizations through the diffusion of external knowledge. Creative
learning involves creating new knowledge and occurs through experimenting, exploration
and creativity.
With this theoretical exertion, I have tried to both thwart the various biases as well
as to borrow from the various interesting insights that already exist within the extensive
literature on organizational learning. In this manner, an attempt was made to integrate the
existing literature on learning and consequently to provide more clarity surrounding the
concept of organizational learning.
185
In part three of the thesis, the implications for organizations was treated. Chapter
eight addressed the question how organizational learning can result in positive outcomes
such as improvement, intelligence and innovation. Given that organizational learning
processes often go astray, I approached the question by addressing the causes of imperfect
learning and how organizations can try to avoid these causes. It was argued that
organizations can strive for successful outcomes of learning by circumventing instances of
focussed selection and interpretation of information - 'focussed learning' - and by
balancing the four types of learning.
Focussed learning can be the result of the dominance of reference groups, selfreferential forces, physical and cultural conditions, specialization, and the occurance of
hidden learning. It results in a lack of diversity within organizations and consequently in
the occurrence of path dependency. Avoiding its occurrence requires more communication
and self-awareness. Dominance of reference groups and self-referential forces often
require a third actor who is able to intervene and analyze imperfect learning processes
from an outsider's point of view. Revealing hidden learning practices can be realized
through the use of ethnographic research studies.
The second general cause of learning imperfections is the occurrence of unbalanced
learning. By unbalanced learning I referred to the situation in which the organization is
focussed too much on one type of learning while ignoring others. Consequently,
organizations should try to complement present learning process with elements of other
types of learning. First of all, this is needed to avoid path dependency. For instance,
internal learning should be complemented by feedback learning in order to detect and
correct errors; without feedback learning, an organization acts as a closed system. Aside
from avoiding path dependency, balancing four types of learning is also necessary to avoid
other negative tendencies. For example, when an organization focusses too much on the
experience of other organizations without reacting to feedback from its immediate
environment, it may fall into the trap of needless mimicking. On the other hand, when an
organization is too focussed on their feedback information, it may fall into a competency
trap. These mutual dependencies exist for all possible combinations of learning types.
In addition to these general implications for organizations which might show them
how to avoid learning defects and improve their learning capacity, organizations can also
make use of information systems as tools to support their learning processes. In chapter
186
nine, I analyzed six information intensive processes that characterizes organizational
learning, in relation to the role of (computerized) information systems: externalization,
objectivation, internalization, selection, interpretation, and idea generation.
It was argued that externalization of knowledge can be supported by the use of
knowledge management systems such as Intranet and Lotus Notes. The discussion was
mainly focussed on the limitations of these systems.
Information systems that support objectivation of knowledge are so called
Organizational Memory Information Systems (OMIS). Again, some marginal boundaries
regarding their effectiveness were drawn.
During the process of internalizing organizational knowledge or memory,
information systems play a role in the spreading of cultural assumptions. As such,
information systems may serve as systems of manipulation.
Selection of knowledge and information may also influence the success of learning.
As argued in chapter eight, inefficiencies of learning are often a result of focussed
information selection. It was argued that most computerized information systems stimulate
this focussed way of learning.
Interpretation of knowledge and information can also influence the success of
learning. While different interpretations may bring about processes of learning under
ambiguity, shared interpretations may result in path dependency. What is needed therefore
are information systems that enhance the sharing of different interpretations. Groupware
systems may be suitable for these purposes.
Finally, information systems that support idea generation are not only important in
supporting creative learning, they also have the potential of avoiding focussed selection
and interpretation of information. The possibilities of simulation software and Groupware
systems as the most accepted information systems that support idea generation, were
discussed. It was argued that unconditioned exploration calls for alternative systems.
These so called Alien Information Systems (AIS) consist of information whose relevance
cannot be determined beforehand and which consequently stimulates the occurance of
chance and serendipity. Information systems which belong to this category of systems are
for example informal networks, libraries, Intranet, and Internet.
The chapter concluded with expressing the need for more research on the
possibilities of information systems to support successful outcomes of learning.
187
10.3
POST-MODERNISM AND ORGANIZATIONAL LEARNING
One could argue that the ideas put forward in this thesis have a rather post-
modernistic touch. In modernistic thinking, aspects such as unity, reference, control and
progress are some of the guiding principles embedded within ideas of 'modernist'
organizations. The dominance of post-modernistic thinking within organizational analysis
has given rise to almost opposing alternative avenues of approach toward the analysis of
organizations and organizing.
Literature on post-modernism and organizations can be divided into two
perspectives. First, there is the perspective that focusses on post-modern organizations
among which are virtual organizations such as network organizations. In this line of
thought, the key is to imagine forms of organizations that can be considered as alternatives
to modernistic forms such as described by Morgan (1986) in his construct of organizations
as machines, as organisms, and to a lesser extent as systems of power. A well-known
representative of this perspective is Clegg (1990).
The other perspective focusses on post-modernistic analysis of organizations where
the emphasis is on the production of organization more than on the organization of
production. Well known representatives of this perspective are Cooper and Burrell (1988).
Here I am more interested in the second approach to post modernism.
Below some of these post-modernist issues are introduced and related to the present
discussion on organizational learning. The reader who is interested in a more substantial
account of post-modernism in relation to organizations, is referred to (Burrell 1988, Clegg
1990, Cooper 1987, 1989, Cooper and Burrell 1988, Gergen 1991, Parker 1992, Power
1990).
First of all, post modernism rejects the idea of an independent, absolute truth.
Modernists would argue that the development of knowledge is not an effort to gain more
and more knowledge about reality. In line with this train of thought, learning cannot be
considered a process of becoming wiser and wiser. Improvement and efficiency gained
through rational strategies such as prescribed for example by the literature on strategic
188
management, information planning, and accountancy, are considered fallacies.
Events and actions are always open to multiple interpretations. These so called
96
'texts' create a social reality that is dominant at that particular time .
Post modernists reject the concept of reference as a univocal relationship between
forms of representations such as words or research and the external world. Instead of
perceiving the knower as independent, or at least partially independent of the knowledge
he or she acquires, post modernists consider the knower and the known as ultimately
connected. This idea is similar to that of self-reference. What we perceive as relevant
knowledge cannot be separated from who we are. "Believing is seeing" is just as valid as
"seeing is believing".
Denying the existence of an absolute truth also implies a rejection of the
modernistic obsession with progress. This is also reflected in the conception of
organizational learning as a process whose outcome, be it progression or deterioration,
cannot be predicted without a close look at the actual processes that make up learning.
The research on organizational learning too can be analyzed from a modernist vs.
post modernist perspective. From a modernist point of view the multiplicity of ways in
which organizational learning is conceptualized and used can be considered a roadblock to
progress. From a post-modern view however, the presence of multiple representations of
organizational learning is not a problem to be solved - it is the way discourses are
(Thatchenkery 1996).
The idea that there is not one absolute truth, nor one absolute organizational
learning model was also present in this thesis. The purpose of the research was to
contribute to the ongoing discourse rather than to present any single theory on
organizational learning.
Related to the idea that there is no single truth is the notion that knowledge is
always socially constructed. This does not imply that human beings are free players.
96
While the interpretation given to a social reality is considered a text, a theory of this reality should be
considered as a story or a narrative, while a meta-theory such as marxism is labeled a meta-narrative.
Likewise, this thesis is nothing more than a story with which you may feel comfortable with or not. To say
that this story is incorrect is within the spirit of the post modernism a fallacy.
189
Rather the opposite, institutional conditions and social context may turn an organization
into a prison (Foucault 1977). Organizational members cannot simply step outside the
institution, since the social world is constructed by it. "(w)e are members whether we like
it or not. All organization does this, all organizations are "total" in the sense that the
prison is what gives us our identity" (Parker 1992, p. 6). Furthermore, in the spirit of
Freud, post modernists argue that individuals do not have their own independent
autonomous identities. There is always 'the other' within the individual.
The idea that the discipline of organizations constructs the individuals within them
is also referred to in this thesis. By focussing on the process of institutionalization and in
specific internalization, I argued that individuals learn to become organizational members.
The AZ case illustrated the dominance of the organizational routines. It also illustrated the
importance of the influence of reference groups during learning. This observation too,
questions the idea of voluntaristic action and free agents.
Post modernists reject the popularity of system thinking within modernistic
accounts on organizations (and other so called 'systems'). Modernists perceive
organizations as systems, where every element within their boundaries is ultimately
connected to other elements, and where environment determines the action of the
organization. In contrast, post-modernists do not see organizations as driving on
determinacy. Chance, surprise and unpredictability are important aspects that trigger
action and make system-thinking irrelevant.
This thesis also goes beyond system-thinking. It has been theoretically argued and
empirically illustrated that the triggers to learning can often be found in for example
serendipitous findings and unexpected events.
Post modernism rejects the obsession of modernism with unity. Organizations are
considered scenes of potential instability. There is always a tension between unity and
diversity, consensus and disagreement. And as Lyotard (1984) argues, this tension is the
energy of social life. In relation to organizations this implies that post modern
organizations incorporate 'difference' through the increase of internal heterogeneity, hiring
of minorities, embracing playfulness, inviting criticism, and enabling external realities to
enter the organization.
Emphasis on difference is mainly reflected in the discussion of 'creative learning'
190
and the way information systems may promote learning. Instead of depending on
information systems that reduce uncertainty, organizational learning, and specifically
learning from others and creative learning, aim toward systems which reduce certainty
and that introduces alien information.
Furthermore, in this thesis, I have stressed the importance of incorporating various
elements - often considered as opposite - such as balancing internal learning with creative
learning. Conditions which foster successful learning also give rise to potential instability.
In addition, behind the ideas presented in this thesis lies the general notion that
organizational behavior is often irrational. Depending on the process of learning, this
irrationality should either be repressed or encouraged.
Rationality is often impeded for example because people learn from experience
which is confounded by experiences of others; or because they are unable to have a
complete picture of their environments and remain unable to see how environments react
on actions of the organization.
These
irrationalities
may
generate
inconsistencies
and
ineffectiveness.
Organizations can strive for more rational behavior. They can create certain conditions,
such as encouraging self-awareness, communication, and diversity of viewpoints.
Paradoxically, pure rationality may also thwart successful learning. Organizational
learning sometimes flourishes through unexpected encounters, foolish behavior, idiot
thoughts, chance events, serendipitous findings, etc. This is especially true for learning
from others and even more so for creative learning.
Given that all four types of learning depend on each other, organizations are faced
with the challenge of balancing rationality with irrationality.
Surely, explicitly approaching organizational learning from a post modernistic
viewpoint requires more than what has been offered in the present thesis. It is my
impression that organizational learning and in particular creative learning, could be
interesting subjects of future post modernistic analysis.
10.4
CONTRIBUTION TO EXISTING KNOWLEDGE
In a oft cited article, Huber (1991) describes and criticizes some of the literature
on aspects of organizational learning while frequently referring to the need for further
191
empirical research. Some of these knowledge gaps have been addressed by the two
empirical studies discussed in this thesis.
For example, he observes that March and Olsen (1976) are two of the few
researchers who describe instances of unintentional or unsystematic learning.
"Other than these retrospective interpretations, there appear to be few if any
published observational or archival studies where unintentional or unsystematic
organizational learning was the focal topic of interest. Systematic field studies of
unintentional
organizational
learning
would
considerably
enhance
our
understanding of the phenomenon and could serve as bases for critiquing an
guiding laboratory and analytical work" (Ruber 1991, p. 94).
Both the AZ case and the case of Lease Co reveal unintentional and unsystematic
learning. These processes of learning could only have been analyzed by studying the
processes at the moment they occurred.
Huber also refers to the need of empirical studies of knowledge acquisition through
"grafting on new members who possess knowledge not previously available within the
organization".
Again, the AZ case provides a vivid example of this specific aspect of learning.
Furthermore, Huber observes that:
"(i)n spite of the importance of organizational experiments as learning
mechanisms... the literature contains very few studies of experimentation by
organizations. What antecedent conditions favor or lead to organizational
experiments? High trust? High need for performance? A culture where tolerance
for mistakes is central?" (Huber 1991, p. 92)
These aspects of learning have been addressed by the introduction of the Lease Co
case.
More fundamental and less focussed on empirical research, Huber concludes his
critical review in the following way:
"A number of conclusions follow from this examination of organizational
learning. One is that the organizational processes and subprocesses that
192
contribute to changes in the range of an organization's potential behaviors
are more numerous and varied than a small sampling of the organizational
science literature might suggest. While any one research group can ignore
this fact with little peril to itself the field as a whole cannot. A second
conclusion is that, with a few exceptions, there is little in the way of
substantiated theory concerning organizational learning and there is
considerable need and opportunity to fill in the many gaps.
A third conclusion flies in the face of the normal science paradigm
and contributes to the just-noted lack of substantiated theory - the
researchers who have studied organizational learning apparently have, to a
surprising degree, not used the results from previous research to design or
interpret their own research. Another conclusion, also contrary to the
advise that scientists frequently give to each other, is that there is little
cross-fertilization or synthesis of work done by different research groups or
on different but related aspects of organizational learning. (An exception to
this conclusion is that James G. March had made important contributions in
a number of areas and has provided a number of integrative works.)"
(Huber 1991, p. 108).
Although I do not think this thesis has (and could have) provided the substantiated
and integrated theory on organizational learning, it is hopefully a step forward in that
direction.
The thesis integrates the various and often distinct perspectives on
organizational learning. Because every perspective on organizational learning can be
criticized for having one or more hidden assumptions or biases that assure an unnecessary
focus on certain directions while overlooking others, a pure integration of the diverse
literature on organizational learning did not suffice. Consequently, I tried to reshuffle the
conventional ideas on organizational learning and mix them with theories from other
disciplines and fields of interests among which theories on innovation, social constructivist
theories, theories on social psychology, theories on institutionalism. The end product of
this mixture was further influenced by ideas that emerged out of the two empirical
research projects, by personal and vicarious experiences, as well as by the substantial
work of March and his friends on organizational behavior.
193
10.5
RECOMMENDATIONS FOR FURTHER RESEARCH
This present thesis should be considered as an exploratory study of organizational
learning. In particular, the aim has been to explore what the concept of organizational
learning means, how it occurs in practice, what the possibilities are to promote it and
what we can expect of the role of information systems during learning. All these questions
had not yet been treated sufficiently in a rigorous scientific manner; there was
consequently a gap in theoretical understanding. However, it was not my aim to fill this
gap completely with scientific evidence. Given that the concept is at the frontiers of
theoretical understanding, a first step towards gathering scientific evidence about it is to
explore its nature and being. This is what I hope to have achieved with the present thesis.
According to standard research practice, the next step in establishing scientific
evidence is to test the ideas that emerged from this exploratory effort. As an attempt to
come together with the scientific community, several ideas introduced in this thesis will be
rewritten as questions for further research.
Of course, there is also another reason why further research is needed. I cannot
claim that this exploratory study covers all subjects at the same level of detail. It is clear
that choices had to be made. For reasons of space and time, but certainly also for reasons
of ignorance, various interesting territoria have been left out. These domains also merit a
cursory glance.
Above all, more empirical research on the processes through which organizations
learn is needed. Given the often unplanned, unsystematic, irrational, and unintentional
nature of learning, this research can only yield interesting findings when it is of a
qualitative, longitudinal nature.
Detailed research of this kind may provide more knowledge about the processes of
internalization and externalization, processes of interpreting feedback
information,
processes of imitation, processes of creating new knowledge, and about the actual process
of balancing the four types of learning. Given the central role of human beings and their
private individual beliefs during these processes of learning, participant observation and
other forms of ethnographic organizational studies are the most suitable to conduct this
kind of research.
Empirical research is further needed on the effects of the use of knowledge
194
management systems on the learning capacity of organizations.
What are the specific requirements of such systems?
Under what conditions are people willing and able to externalize their private
knowledge?
How do we up-date these systems?
What are the unintended consequences of the implementation of knowledge
management (systems)?
More empirical research is needed on the role of Internet and World Wide Web
during organizational learning; it would be particularly useful in studying 'learning from
others' and 'creative learning'. World Wide Web has just recently entered the field of or­
ganizations. Although organizations have certainly gained learning-experience while using
the system, results of this learning has not yet been reported, at least not in the field of
information systems' or organization studies. Given this relatively new territory, research
calls for empirical exploration, such as qualitative (ethnographic) case studies.
Although the interest in organizational memory is growing, there is still no clear
understanding of this concept. The construct of organizational memory is clearly important
to the idea of organizational learning, but has studied very infrequently. Consequently,
just as organizational learning called for an exploratory study, the same applies to the
concept of 'organizational memory'.
Again, just as is the case for organizational learning, organizational memory can
be considered as a generic phenomenon. Every organization has at least one, but probably
multiple, memories. As Huber (1991) already advocated half a decade ago, but what is
still not well understood, is the extent to which non-routine information is deliberately
stored to be used as a basis for future learning practices.
There is an interesting area that has not yet been covered by past or present
research. This concerns the research question: "Why has organizational learning become
so popular at the moment while its genesis lies somewhere around the turn of the
century?". Is the rise of a 'knowledge society' and the increasing turbulance of
environments, the only explanation for this popularity or are there more? The same
question can be posed with respect to the sudden rise in popularity of the notion of
195
'knowledge management'.
Related to this is the question of the 'price of learning'. Whereas internal learning,
feedback learning and to a lesser extent learning from others are frequently occurring
organizational processes, creative learning is not (March 1991). This is predominantly so
because of the inherent conservatism of organizations; creativity seems a difficult goal to
achieve. (March 1991). Given this difficulty in fostering creative learning, it would be
interesting to study the returns of learning. We could think for example of time, attention,
risks and organizational resources (people, money, goodwill), as prices to be paid when
organizations want to engage in learning. When we set this against the perennial ups and
downs of business cycles (Perez 1985), addressing this issue may reveal interesting ideas
related to the question of the present popularity of organizational learning.
Several organizational issues are currently the subject of intense discussion, such as
outsourcing, virtual organizations, flexible work, teleworking, Total Quality Management,
Business Process Redesign, to name but a few. They all have their implications for
organizational learning. I have considered briefly some of these issues but they certainly
need more research attention. For example:
To what extent are organizations able to construct organizational knowledge when
the organizational members only interact with each other through IT media such as
e-mail?
Given the focus on short term efficiency, what are the implications of Total
Quality Management techniques for organizational learning?
What are the implications of outsourcing for the learning capacity of the
organization and what does this imply in the long run?
Under what conditions do Business Process Redesign-efforts result in successful
reformations or are its claims of radicalism and novelty indeed exaggerated (Grint
et al 1995)?
More research is needed on the possibilities and potentials of doing ethnographic
research to inform organizational practitioners about the actual organizational learning
processes that occur within organizations.
More conceptual work is needed on the relation between organizational learning
and post modernism, and in particular on the issue of creative learning and the use of
196
Alien Information systems.
Finally, more technically oriented research is needed on the design of information
systems that support organizational learning processes.
In general, I hope this present thesis will spur future research studies to treat
organizational learning as a process rather than an outcome. Researchers should spend
more effort in studying how improvement and intelligence can be promoted by means of
learning processes.
197
APPENDIX I
ASSESSING ORGANIZATIONAL LEARNING
CAPACITY
1)
Is there a firm awareness of the actions taking place within the organization?
2)
Is there an understanding of the reasons for these actions?
3)
To what extent are individual beliefs related to the organizational knowledge?
4)
To what extent does the organization also respond to environmental reactions?
5)
Are these reactions based on organizational actions?
6)
Are these reactions interpreted as intended by the environment?
7)
Does the organization also make use of the experience of other organizations?
8)
To what extent does this external knowledge correspond to existing knowledge?
9)
Does the organization consider alternative models to imitate?
10)
Does the organization also allow for experimenting and creativity without wishing
to achieve results in the short run?
11)
Is there enough internal communication to spread the experience of creative
learning among organizational members?
ad 1) The first question refers to the possible occurrence of audience learning as
mentioned in chapter four and in section 8.2. Managers are often unaware of actual
learning processes which are taking place within the organization. Consequently, the first
step in assessing current learning capacity of the organization is to understand the actual
learning processes. This requires close contact with the various communities of practices
that exist within the organization. In case audience learning occurs, organizations and
managers should ask themselves why the organization allows for audience learning and
how the level of internal communication can be improved.
ad 2) In inquiring whether managers understand the reasons behind individual actions, I
refer to the possible occurrence of restrained learning. Restrained learning occurs in those
cases in which individual learning is frustrated. Although individuals draw their own
conclusions based on experience, they cannot put these into practice; they may be
constrained, for example, by their job descriptions dictated by management. Consequently,
being aware of actions within an organization is not enough to understand actual learning
processes. Organizational learning may be hampered if managers overlook the actual
theories-in-use. Again, managers should ask themselves why the organization allows for
restrained learning. Restrained learning will be avoided when organizations stimulate the
communication of private individual beliefs. In the next chapter, I will discuss the use of
information systems to support this information exchange.
ad 3) The question to what extent individual beliefs are related to present organizational
knowledge, refers to the possible occurrence of anarchistic learning. When individual
beliefs are not very much in line with present organizational knowledge, this could have
negative consequences. Individuals may not in fact share their personal beliefs. When
differing views might bring valuable changes to the organization, a certain degree of
anarchistic learning might be helpful. Given that past experiences often dominate, learning
from these different views often requires a carefully monitored learning process. When
assimilation of the different views does not succeed, in other words when mutual learning
is unbalanced, individuals become too much disassociated from the organization to act as
organizational members. In these instances, their learning remains individual learning and
does not contribute to any organizational level of learning. Adjusting training and
200
socialization practices to allow for more mutual learning could be a possible solution to
stimulate anarchistic learning.
4)
Questions one to three referred to internal learning. As argued in section 8.3,
internal learning should be balanced with feedback learning in order to understand the
effects of organizational actions and determine whether actions need to be adjusted to
environmental demands. Question four addresses the degree in which organizations allow
for this feedback information. The first step is to conduct a stakeholder analysis by listing
actors, perceived as relevant, within the environment from which the organization acquires
feedback information. This list would possibly reveal the existence of actors who might be
important but who are never perceived as such.
5)
In deciding whether environmental reactions are based on organizational action, the
possible existence of 'superstitious learning' is addressed. Superstitious learning occurs
when the organization assumes that environmental reactions are based on organizational
action while in fact the environment reacts independently of these actions. In order to
avoid superstitious learning, the organization should have a sound understanding of the
reasons behind environmental reactions. This requires rich communication between the
various stakeholders in the organization.
6)
refers
The next question deals with interpretation of these environmental reactions and
to
learning under
environmental
ambiguity.
(re)actions are
Learning
under
interpreted differently
ambiguity
than was
happens
when
intended by the
environment. A single (re)action may result in multiple explanations. In case organizations
are unaware of the existence of multiple explanations, learning under ambiguity might be
a hinderance toward organizational learning. Consequently, the first step is to understand
its possible occurrence and the reason why it occurred. In itself, multiple interpretations
may contribute to diversity and consequently to a (re)construction of organizational
knowledge. However, in order to benefit from existing multiple interpretations, they
should be shared amongst members in the organization. More or less autonomous units
connected to each other through various communication channels, could support the
existence and sharing of multiple interpretations.
201
7)
Whereas questions five and six referred to feedback learning, the question whether
the organization also makes use of the experience of other organizations refers to
balancing feedback learning with processes of learning from others. Without learning from
the experience of other organizations, organizational learning might result in unintended
conservatism or path dependency. The organization is focussed on the effectiveness of its
status quo while it overlooks possible alternative ways of acting or alternative
environments to respond to.
8)
When organizations do indeed engage in learning from others, it is important to
assure that the external knowledge from which the organization learns corresponds to
some extent to existing organizational knowledge. Attractiveness of ideas, products,
technologies or other innovations is not a guarantee to success. The rapid changes of
innovations within the IT business for example have seduced many IT managers to adopt
them without considering the degree of readiness for them within the organization
(Davenport 1996). The gap between existing knowledge and the new knowledge that the
organization wishes to integrate, should not be too big because in this way implementation
problems can be avoided. Again, this requires self-awareness as well as intense
communication both within and outside the organization.
9)
From time to time, organizations should consider alternative models to imitate with
the aim of discouraging path dependency. Again, a stakeholder analysis consisting of
environmental actors who are presently perceived as relevant and who could be of
relevance in the future, might provide insight in current knowledge-gaps. It is also
important to understand why the organization does not consider alternative sources of
external knowledge to imitate. Information systems that are used to monitor the
environment may be too focussed for example.
10)
The question whether the organization also allows for experimenting and creativity
addresses the issue of balancing creative learning with the previously discussed types of
learning. Of course, the need and possibility to engage in creative learning depends to a
great extent on the slack resources of the organization as well as on the degree in which
the environment is perceived as being uncertain. In most cases however, an absence of
risk-taking, experimenting, playfulness, and creativity, will increase the chance of falling
202
in the trap of path dependency with its negative consequences in the long run. For one,
creative learning is necessary in order for an organization to be or to remain ahead of its
competitors. Dependence on other organizations to gain new knowledge also causes the
organization to confront problems of assimilating the knowledge to the existing knowledge
as discussed under question eight. This can result in even more problematic situations
when we consider the fast changing environments with which most organizations are
coping nowadays. Creating new knowledge through creative learning is also important in
order to create and maintain the capacity to recognize and absorb new external knowledge.
11)
The last question deals with the sharing of the results of creative learning and
refers to the link between creative learning and internal learning. After all, when
outcomes of creative learning are not spread among organizational members, organizations
cannot capitalize on these experiences. This implies that actors engaging in creative
learning should to some extend be integrated within the organization as to enable the
diffusion of experiences among the other organizational members. As mentioned earlier,
this need to communicate experiences is also necessary to enable learning from others,
that is to be able to recognize the value of new, external knowledge, assimilate it, and
apply it.
203
REFERENCES
Abbott, A. (1988) The Systems of professions, Chicago: University of Chicago Press.
Agar, M.H. (1986) Speaking of ethnography, Sage University Paper Series on Qualitative
Research Methods, vol 2, Beverly Hills CA: Sage.
Aldrich, H. and D. Herker (1977) "Boundary spanning roles and organizational structure",
Academy of Management Review, April, 217-230.
Allen, R.C. (1983) "Collective invention", Journal of Economic Behavior and Or­
ganization, vol 4, pp. 1-24.
Ansoff, H.I. (1965) Corporate Strategy, New York: McGraw Hill.
Argyris, Ch. (1977) "Organizational learning and management information systems",
Accounting, Organizations and Society, vol 2/2, pp. 113-123.
Argyris, Ch. (1985) Strategy, change and defensive routines, Boston: Pitnam.
Argyris, Ch. (1990) Overcoming organizational defenses: Facilitating organizational lear­
ning, Needham Heights, MA: Allyn and Bacon.
Argyris, Ch. and D. Schön (191%)) Organizational learning: a theory of action-perspective"
Reading, MA: Addison-Wesley.
Arrow, K.J. (1962) "The economic implications of learning by doing", Review of Economic
Studies, vol 29, pp. 155-173.
Astley, W.G. and A.H. van der Ven (1983) "Central perspectives and debates in organiza­
tion theory", Administrative Science Quarterly, vol 28, pp. 245-273.
Bateson, S. (1973) Steps to an ecology of mind, Frogmore: Paladin.
Beath, C M . (1991) "Supporting the information technology champion", MIS Quarterly,
september, pp. 355-372.
Beninger, J.R. (1986) The control revolution, Cambridge MA: Harvard University Press.
Berger, P. and T. Luckman (1966) The social construction of knowledge, London: Penguin
books.
Blackler, F. (1995) "Knowledge, knowledge work and organizations: An overview and
interpretation", Organization Studies vol 16/6, pp. 1021-1046.
Blonk, van der H. (1996) "Rethinking the concept of information systems", paper presented
at the SWEHOL working conference on "Managing our technical society" Maarssen,
The Netherlands, April, pp. 15-18.
Boersma,
S.K.T.
(1995) Kennismanagement: een creatieve onderneming, (transí.:
Knowledge management, a creative enterprise) Inaugúrale rede, Universiteit
Groningen.
Bohm, D. (1990) "David Bohm on dialogue" David Bohm Seminars, Ojai.
Boland, R.J. (1987) "The in-formation of information systems", In: R.J. Boland (ed)
Critical Issues in Information Research, R.J. Boland.
Boland R.J. (1991) "Information system use as a hermeneutic process", Proceedings of the
IFIP Conference: "Information systems research: Contemporary approaches and
emergent traditions", Elsevier Science Publishers, pp. 439-464.
Boland, R.J, R.V. Tenkasi and D. Te'eni (1994) "Designing information technology to
support distributed cognition", Organization Science, vol 5/3, pp. 456-475.
Bolk, H. (1989) Organizing, changing, simulating, An organizing perspective for manage­
ment, consulting and science policy, Delft: Eberon.
Boogaard, M. (1994) Defusing the software crisis, Information systems flexibility through
data independence, PhD thesis, Amsterdam: Thesis Publishers.
Bordieu, P. (1973) "Cultural reproduction and social reproduction" In: I. Brown (ed)
Knowledge, education and cultural change, London: Tavistock, pp. 71-112.
Brown J.S. and P. Duguid, (1991) "Organizational learning and communities-of-practice:
towards a unified view of working, learning, and innovation", Organizational
Science, vol 2/1, pp. 40-57.
Brunsson, N. (1985) The irrational organization, Chichester: John Wiley.
Brunsson, N. (1989) The organization of hypocrisy, Chichester: John Wiley.
Brunsson, N. and J.P. Olsen (1993) The reforming organization, London: Routledge.
Burgelman, R.A. (1990) "Strategy-making and organizational ecology: A conceptual
framework". In: J.V. Singh (ed) Organizational Evolution, Newbury Park, CA:
Sage publications, pp 164-181.
Burgelman, R.A. and L.R. Sayles (1986) Inside corporate innovation, New York: The Free
Press.
Burns, T. and G.M. Stalker (1961) The management of innovation, London: Tavistock.
Burt, R.S. (1980) "Models of network structure", Annual Review of Sociology, vol 6, pp.
79-141.
Burell, G. (1988) "Modernism, postmodernism and organizational analysis 2: The
contribution of Michel Foucault", Organization studies, vol 9/2, pp. 221-235
Cangelosi, V.E. and W.R. Dill (1965) "Organizational learning: Observations towards a
206
theory", Administrative Science Quarterly, vol 10, pp. 175-203.
Ciborra, C.U. and G.F. Lanzarra (1994) "Formative contexts and information technology,
understanding
the
dynamics
of
innovation
in
organizations",
Accouning,
Management, and Information technology vol 4/2, pp 61-86.
Christiaansen, E. (1994) Strategic advantage and the exploitability of Information
technology, An empirical study of the effects of IT on supplier-distributor
relationships in the US airline industry, PhD thesis, Amsterdam: Thesis Publishers.
Clarke, A.E. (1991) "Social worlds/arenas theory as organizational theory", In: D.R.
Maines (ed) Social organization and social process, Essays in honor of Anselm
Strauss, New York: Aldine de Gruyter.
Clegg, S.R. (1990) Modern organizations, Organization studies in the Postmodern World, M
London: Sage Publications.
Coenen H., P. Pennartz, and F. Wester (ed) (1988) Nasporingen: structuren van
interpretatief sociaal wetenschappelijk onderzoek, (trans: Investigation: structures of
interpretive social scientific research), Amsterdam: Siswo.
Cohen, M.D. and D. Levinthal (1990) "Absoptive Capacity: a new perspective on learning
and innovation", Administrative Science Quarterly, March, pp. 128-152.
Cooper, R. (1987) "Information, communication and organization: a post-structural
revision", Journal of Mind and Behavior, vol 8/3, pp. 395-416.
Cooper, R. (1989) "Modernism, postmodernism and organizational analysis 3: The
contribution of Jacques Derrida", Organization Studies, vol 10/4, pp. 479-502.
Cooper, R. and G. Burrell (1988) "Modernism, postmodernism and organizational analysis:
an introduction", Organization Studies, vol 9/1, pp. 91-112.
Copeland, D.G. and J.L. McKenny (1988) "Airline reservations systems: lessons from
history" MIS Quarterly, September, pp. 353-370.
Crossan, M. and T. Guatto (1996) "Organizational learning research profile", Journal of
Organizational Change Management, vol 9, pp. 107-113.
Cyert, R.M. and J.G. March (1963) A behavioral theory of the firm, Englewood Cliffs, NJ:
Prentice Hall.
Daft, R.L. and G.P. Huber (1987) "How organizations learn: a communication process",
Research in the Sociology of Organizations, Greenwich: JAI-Press, pp. 1-36.
Daft, R.L. and R.H. Lengel (1986) "Organizational information requirements, media
richness and structural design", Management Science, vol 32/5, pp. 54-571.
Daft, R.L., J. Sormunen and D. Park, (1988) "Chief executive scanning, environmental
characteristics and company performance", Strategic Management Journal, vol 9,
pp. 123-139.
Daft, R.L. and D.E. Weick (1984) "Toward a model of organizations as interpretation sys­
tems" Academy of Management Review, vol 9/2, pp. 284-295.
Davenport, Th.H. (1996) "Why reengineering failed, the fad that forgot people", Fast
Company, premier issue, pp. 70-74.
Davis, G.F. and W.W. Powell (1992) "Organization-environment relations" In:
M. Dunnette et al (ed) Handbook of industrial and organizational psychology, vol 3,
second edition, Consulting Psychologists Press.
Dawe, A. (1970) "The two sociologies", Britisch Journal of Sociology, vol. 21, p. 207-218.
De Geus, A.P. (1988) "Planning as learning" Harvard Business Review, March-April.
DeSanctis, G. and B, Gallupe (1987) "Group decision support systems", Management
Science, vol 33/5, pp. 589-609.
DiMaggio, P.J. and W.W. Powell (1983) "The iron cage revisited: Institutional
isomorphism
and
collective
rationality
in organizational
fields", American
Sociological Review, vol 48, pp. 147-160.
Dodgson, M. (1993a) "Learning, trust and technological collaboration", Human Relations,
vol 46, pp. 77-95.
Dodgson, M. (1993b) "Organizational Learning: A Review of Some Literatures", Or­
ganization Studies, vol 14/3, pp. 375-394.
Dongen van, H.J. (1991) "Some notions on social integration and steering". In R.J. i n ' t
Veld et al (ed) Autopoiesis and configuration theory: new approaches to societal
steering, Dordrecht: Kluwer, pp. 47-54.
Downs, A. (1966) Inside bureaucracy, Boston: Little, Brown and Company.
Drucker, P.F. (1988) "The Coming of the New Organization" Harvard Business Review
jan-february, pp. 45-53.
Duncan, R.B. and A. Weiss (1979) "Organizational learning: implications for or­
ganizational design", Research in Organizational Behavior, JAI press, pp. 75-123.
Durkheim, E. (1964) Rules of sociological method, Free Press.
Durkheim, E. (1978) On institutional analysis, Chicago: Chicago University Press.
Dewey, J. (1928) The philosophy of John Dewey, London: Allen & Unwin.
Dijk, Van N. M. H. (1989) Een methodische strategic van organizatieverandering (transl. a
208
methodical strategy of organizational change) Delft: Eburon.
Edel, R. (1996) Service kwaliteit en organisatie bij de Nederlandse Spoorwegen (transí.
service quality and organization at the Netherlands Railways), masters thesis, Vrije
Universiteit Amsterdam, February.
Easterby-Smith, M. (1996) "Disciplines of the learning organization: contributions and
critiques", Proceedings of the Symposium "Organizational learning and the learning
organization, Theoretical and research developments", Lancaster 1-3 September.
Feyerabend, P. (1975) Against methods, London NLB.
Finney, M. and I.I. Mitroff (1986) "Strategic plan failures: the organization as its own
worst enemy" In: D.A. Gioia and H.R. Sims (ed) The thinking organization, Jossey
Bass, San Francisco, pp. 317-335.
Fiol, C M . and M.A. Lyles (1985) "Organizational learning", Academy of management
review, vol 10/4, pp. 803-813.
Forester, J. (1961) Industrial dynamics, Cambridge: MA, MIT Press.
Forssell, A. (1989) "How to become modern and businesslike: an attempt to understand the
modernization of Swedish savings banks", International Studies of Management and
Organization, vol 19/3, pp. 34-48.
Foskett, D.J. (1984) Pathways for communication, London: Clive Bingley.
Foucault, M. (1977) Discipline and punish: the birth of the prison. London: Allen Lane.
Fung, Y. (1948) A short history of Chinese philosophy, New York: The Free Press.
Fukuyama, F. (1995) Trust: the social virtues and the creation of prosperity, New YorK:
Free Press.
Galbraith, J.R. (1973) Designing complex organizations, Reading, MA: Addison-Wesly.
Galbraith, J.R. (1982) "Designing the innovating organization", Organizational Dynamics,
vol. 11, pp. 5-25.
Garrat, B. (1987) The learning organization, Worcester: Billings.
Garvin, D.A. (1993) Building a learning organization, Harvard Business Review, vol 71/4,
pp. 78-91.
Genschel, P. (1997) "The dynamics of inertia: institutional persistence and change in
telecommunications and health care", to be published in Governance, vol 10/1.
Gergen, K. (1991) "Organizational theory in the postmodern era", In: M. Reed (ed) New
directions in organization theory and analysis, London, Sage.
Giddens, A. (1976) New rules of sociological method, London: Hutchinson.
209
Giddens, A. (1979) Central problems in social theory, Berkeley: University of CA. Press.
Giddens, A. (1984) The Constitution of Society, Polity Press, Cambridge.
Gluck, F.W. (1985) "Big bang management; creative innovation", Administrative Science
Quarterly, spring, pp. 49-59.
Glaser, B.G. and A. Strauss (1976) The discovery of grounded theory, Chicago: Aldine.
Goffman, E. (1959) The presentation of self in everyday life, Garden City, New York:
Doubleday Anchor.
Goffman, E. (1983) "The interaction order" American Sociological Review, vol 48, pp. 117.
Granovetter, M. (1973) "The strength of weak ties", American Journal of Sociology, vol
78, pp. 1360-1380.
Grint, K., P. Case, and L. Wilcocks (1995) "Business process reengineerig reappraised: the
politics and technology of forgetting". In: Information Technology and changes in
organizational work, W.J. Orlikowski et al (ed.), London, Chapman & Hall 39-61.
Halberstam, D. (1972) The best and the brightest, Berrie & Jenkins: London.
Hakansson, H. (1987) Industrial technological development: a network approach, London:
Croom Helm.
Hammer, M. and J. Champy (1993) Reengineering the corporation: a manifesto for
business revolution, Harper Business, New York.
Hammersley, M. and P. Atkinson (1983) Ethnography, principles in practice, London:
Routhledge.
Handy, C. (1995) "Work and organisations: a profile of Charles Handy, writer and
educator". An interview with Paul Stonham", European Management Journal vol
13/4, pp. 374-381.
Handy, C. (1995) "Trust and the virtual organization", Harvard Business Review, mayjune, pp. 40-50.
Hedberg, B.L.T. (1981) "How organizations learn and unlearn". In: P.C. Nystrom, and
Starbuck, W.H. (ed) Handbook of Organization Design, vol 1, pp. 3-27.
Hedberg, B. and S. Jonsson, (1978) "Designing semi-confusing information systems for
organizations in changing environment", Accounting, Organization and Society, vol
3/1, pp. 47-64.
Hedlund, G. and I. Nonaka (1993) "Models of Knowledge management in the west and
Japan". In: P. Lorange et al (ed) Implementing strategic processes, change, learning
210
and co-operation, Oxford: Basil Blackwell.
Heng, M.S.H. and I.S.Y. Koh (1992) Towards a wider scope of using information in
organization, Vrije Universiteit: Serie Research Memoranda.
Heng, M.S.H., E.M. Trauth, and S.J. Fischer (1994) An innovation model of information
planning, Vrije Universiteit: Serie Research Memoranda.
Herriot, S.R., D. Levinthal, and J.G. March (1988) "Learning from experience in
organizations". In: J.G. March Decisions and organizations Oxford, Basic Blackwell
Ltd., pp. 219-227.
Hippel von, E. (1988) The Sources of Innovation Oxford: Oxford University Press.
Hirschman, A.O. (1970) Exit, voice and loyalty: Responses to decline in firms,
organizations, and states, Cambridge: Harvard University Press.
Hirschhorn, L. (1984) Beyond mechanization, work and technology in a postindustrial age,
Cambride MA: MIT Press.
Hofstede, G. (1980) Culture's consequences: international differences in work related
values, Beverly Hills, CA: Sage.
Hopstaken, B.A.A. and A. Kranendonk (1990) Informatieplanning - puzzelen met beleid en
plan (transl: Information planning - puzzling with policy and plan)" Leiden: Stenfert
Kroese.
Hosking, D.M. and I.E. Morley (1991) A social psychology of organizing, New York:
Harvester Wheatsheaf.
Huber, G.P. (1984) "The nature and design of post-industrial organizations", Management
Science, vol 30/8, pp. 928-951.
Huber, G.P. (1991) "Organizational learning: the contributing processes and the
literatures", Organizational Science, vol 2/1, pp. 88-115.
Hutchins, E. (1991) "Organizing work by adaptation", Organization Science, vol 2/1, pp.
14-39.
Huysman, M.H. (1992) Different perspectives on organizational learning, a literature
survey, Vrije Universiteit: Serie Research Memorandum.
Huysman, M.H., S.J. Fischer, and M.S.H. Heng (1993) Images of organizations and
organizational learning, implications for the use of information. Unpublished
research paper, Vrije universiteit, Amsterdam.
Huysman, M.H., S.J. Fischer and M.S.H. Heng (1994) "An organizational learning
perspective on information systems planning", Journal of Strategic Information
211
Systems, vol 3/3, pp. 165-177.
Huysman, M.H., R. Frisart and M.S.H. Heng (1995) "Autopoiesis and organizational
learning, implications for information systems", Proceedings SISnet conference,
Bern
Huysman, M.H. (1996a) "Dynamics of mutual learning", Scandinavian journal of
management, under review.
Huysman, M.H. (1996b) "Creating the Mobility Pass, A case of generative learning",
Proceedings European Conference on Information Systems, Lisbon.
Isaacs, W. (1993) "Taking flight: dialogue, collective thinking and organizational learning",
Organizational Dynamics Fall.
Isaacs, W. and P. Senge (1992) "Overcoming limits to learning in computer-based learning
environment" European Journal of Operational Research, vol 59, pp. 183-196.
James, W. (1950) The principles of psychology, New York: Dover.
Janis, I. L. (1972) Victims of group think, Boston MA: Houghtew-Mifflin.
Jelinek, M. (1979) Institutionalizing innovations, a study of organizational learning systems,
New York: Praeger.
Kanter, R.M. (1977) Men and women of the corporation, New York: Basic Books.
Kanter, R.M. (1983) The change masters, New York: Simon and Schuster.
Kanter, R.M. (1988) "When a thousand flowers bloom: structural, collective, and social
conditions for innovation in organization", Research in Organizational Behavior, vol
10, pp. 169-211.
Kim, D.H. (1993) "The link between individual and organizational learning", Sloan
Management Review, Fall, pp. 37-50.
Kimberly, J.R. (1981) "Managerial innovation". In: P. Nystrom and W. Starbuck Handb­
ook of Organizational Design Vol 1, London: Oxford University Press.
Kirckpatrick, D. (1996) "IBM and Lotus: not so dumb after all", Fortune, July 8, pp. 3438.
Kirk, J. and M. L. Miller (1986) Reliability and validity in qualitative research, Sage
University Paper Series on Qualitative Research Methods, vol 1, Beverly Hills, CA:
Sage.
Klumper, D.J. (1996) The history of organizational learning, masters thesis, Vrije
Universiteit Amsterdam.
Koestler, A. (1964) The act of creation, London: Arkana.
212
Kreiner, K. and M. Schultz (1990) "Informal collaboration in R&D, the formation of
networks across organizations" Organization Studies, vol 14/2, pp 189-209.
Kuhn, T.S. (1970) The structure of scientific revolution, second edition, Chicago: Univer­
sity of Chicago Press.
Latour, B. (1987) Science in action: How to follow scientists and engineers through society,
Milton Keynes: Open university press.
Lave J. and E. Wenger (1991) Situated learning: legitimate peripheral participation,
Cambridge, UK: Cambridge University Press.
Lawrence, P.R. and J.W. Lorsch (1967) Organization and environment, Cambridge MA:
Harvard Graduate School of Business Administration.
Lazarsfeld, P.F. and R.K. Merton (1964) "Friendship as social process: a substantive and
methodological analysis". In: M. Berger et al (ed) Freedom and control in modern
society, New York: Octagon.
Lee, A.S. (1994) "Electronic mail as a medium for rich communication: an empirical
investigation using hermeneutic interpretation", MIS quarterly, June, pp 143 - 157.
Leonard-Barton D. (1987) "The case for integrative innovation: an expert system at
Digital", Sloan Management Review, Fall, pp. 7-19.
Leifer, R. and G.P. Huber, (1977) "Relations among perceived environmental uncertainty,
organizational structure, and boundary spanning", Administrative Science Quarterly,
vol 22, pp. 235-247.
Levinthal, D. and J.G. March (1993) "The myopia of learning", Strategic Management
Journal, vol 14, pp. 95-112.
Levi-Strauss, C. (1966) The savage mind, Chicago: Chicago University Press.
Levitt, B. and J.G. March (1988) "Organizational learning", Annual Review Sociology, 14,
pp. 319-340.
Lewin, K. (1957) "Frontiers in group dynamics: concept, method, and realily in social
science", Human Relations, vol 1, pp. 5-42.
Lounamaa, P.H. and J.G. March (1987) "Adaptive Coordination of a Learning Team"
Management Science, vol 33, pp. 107-123.
Lowery, S.A. and M.L. DeFleur (1988) Milestones in mass communication research: media
effects, second edition, New York: Longman.
Lyotard, J.F. (1984) The postmodern condition: a report on knowledge, Manchester:
Manchester University Press.
213
Maas, A.J.J.A. (1988) Ongedefinieerde mimten (transí. Undefined social spaces) PhD
thesis, Delft: Eburon.
MacKenzie D. and J. Wajcman (1985) The social shaping of technology Milton Keynes
Open University Press.
Maidique M.A. (1980) "Entrepreneurs, champions and technological innovation" Sloan
Management Review winter, pp. 59-76.
March, J.C. and J.G. March (1977) Almost random careers: the Wisconsin school
superintendency 1940-1972. Administrative Science Quarterly, vol 22, pp. 377-409.
March, J.G. (1988) "Technology of foolishness" In: J.G. March Decisions and
Organizations, Oxford: Basic Blackwell.
March, J.G. (1988) Decisions and Organizations, Oxford: Basil Blackwell.
March, J.G. (1990) "The evolution of evolution". In: J. V. Singh (ed) Organizational
evolution: new directions, Newbury Park, CA: Sage.
March, J.G. (1991) "Exploration and exploitation in organizational learning", Or­
ganizational Science, vol 2/1, pp. 71-87.
March, J.G. (1994) A Primer on decision making, New York: Free Press.
March, J.G. and J.P. Olsen (1976) Ambiguity and choice in organizations, Bergen,
Norway: Universitetsforlaget.
March, J.G. and J.P. Olsen (1989) Rediscovering institutions, New York: Free Press.
March, J.G. and G. Sevon (1984) "Gossip, information, and decision-making", Advances in
Information Processing in Organizations, vol 1, pp. 95-107.
March, J.G. and Z. Shapira (1987) "Managerial perspectives on risk and risk taking",
Management Science, vol 33/11, pp. 1404-1418.
March, J.G., L.S. Sproull, and M. Tamuz, (1991) "Learning from samples of one or
fewer", Organizational Science, vol 2/1, pp. 1-13.
March, J.G. and H.A. Simon (1993) Organizations, second edition, Cambridge MA:
Blackwell Publishers.
Maso, I. (1988) Kwalitatief onderzoek (transí. Qualitative Research), Meppel: Boom.
Mason, R.O. and I.I. Mitroff (1981) Challenging Strategic Planning Assumptions, Los
Angeles: John Wiley & Sons.
Maturana, H.R. and F.J. Várela (1980) Autopoiesis and Cognition: the realization of the
living, Dordrecht: Reidel.
Mayer, R.C., J.H. Davis, and F.D. Schoorman (1995) "An integrative model of
214
organizational trust", Academy of Management Review vol. 20, pp. 709-734.
McGregor, D. (1960) The human side of enterprise, New York: McGraw-Hill.
McKee, D. (1992) "An organizational learning approach to product innovation" Journal of
Product Innovation Management, vol 9, pp. 232-245.
McPerson, J.M., P.A. Popielarz, and S. Drobnic (1992) "Social networks and
organizational dynamics", American Sociological Review, vol 57, pp 153-170.
Mead, H. (1934) Mind, Self and Society, Chicago: University of Chicago Press.
Meyer, J.W. and B. Rowan (1977) "Instititualized organizations: formal structure as myth
and ceremony", American Journal of Sociology vol. 83, pp. 340-363.
Miles, R.H. and W.A. Randolf, (1980) "Influence of organizational learning styles on early
development", In: J.R. Kimberly and R.H. Miles (ed) The organizational life cycle:
issues in the creation, transformation, and decline of organizations, San Francisco:
Jossey-Bass, pp. 44-82.
Miller, D. and P.H. Friesen, (1980) "Momentum and revolution in organizational adap­
tation" Academy of Management Journal, vol 23/4, pp. 591-614.
Miller, D. (1994) "What happens after success: the perils of excellence", Journal of
Management Studies, vol 31/3, pp. 325-358.
Milliken, F.J. and L.L. Martins (1996) "Searching for common threads: understanding the
multiple effects of diversity in organizational groups" The Academy of Management
Review, vol 21/2, pp. 402-433.
Mintzberg, H. (1979) The structuring of organizations, Englewood Cliffs, NJ: Prentice
Hall.
Mintzberg, H. (1983) Structures in Five, designing effective organizations, Englewood
Cliffs, NJ: Prentice hall.
Mintzberg, H. (1988) "Opening up the definition of strategy", In: J.B. Quinn et al (ed) The
strategy process; concepts, contexts and cases, Englewood Cliffs, N.J.: PrenticeHall, pp. 13-20.
Mintzberg, H. (1989) Inside our strange world of organizations, New York: Free Press.
Mintzberg, H. and J.A. Waters (1985) "Of strategies, deliberate and emergent", Strategic
Management Journal, vol 6, pp. 257-272.
Mirvis, P.H. (1996) "Historical foundations of organizational learning", Journal of
Organizational Change, vol 9/1, pp. 13-31.
Moingean, B. and A. Edmondson (1996) "Trust and organizational learning", Proceedings
of the Symposium "Organizational learning and the learning organization,
Theoretical and research developments", Lancaster 1-3 september.
Morgan, G. (1986) Images of Organization, Newsbury Park, CA: Sage Publications.
Morgan, G. and R. Ramirez (1983) "Action learning: a holographic metaphor for guiding
social change, Human Relations, vol 7/1, pp. 1-28.
Nagasundaram, M. and R.P. Bostrom (1994) "The structuring of creative processes using
GSS: A framework for research", Journal of Management Information Systems, vol
11/3, pp. 87-114.
Nelson, R. and S.G. Winter (1982) An evolutionary theory of economic change,
Cambridge, MA: Addison-Wesley.
Nohria, N. and R. Eccles (ed) (1991) Networks and organizations, Boston: Harvard
Business School Press.
Nonaka, I. (1988) "The knowledge-creating company" Harvard Business Review, nov-dec,
pp. 96-104.
Nonaka, I. (1990) "Redundant, overlapping organization: a Japanese approach to managing
the innovation process" California Management Review, Spring, pp. 27-38.
Nonaka, I. and J.K. Johansson (1985) "Organizational learning in Japanese companese",
Advances in Strategic Management, vol 3, pp. 277-296.
Nunamaker, J.F., L.M. Applegate, and B.R. Konsynski (1987) "Facilitating group
creativity: experience with a group decision support system", Journal of
Management Information Systems, vol 3/4, pp 5-19.
O'Reilly, C A . , D.F. Caldwell, and W.P. Barnett (1989) "Work group demography, social
integration and turnover", Admistrative Science Quarterly, vol. 34, pp. 21-37.
Orlikowski, W.J. and D. Robey (1991) "Information technology and the structuring of
organizations" Information Systems Research, vol. 2/2, pp. 143-169.
Orr, J. E. (1990) "Sharing knowledge, celebrating identity: Community memory in a
service culture", In: Middleton, P. and D. Edwards et al (ed) Collective
Remembering, Memory in society London: Sage Publications.
Parker, M. (1992) "Post modern organizations or Postmodern organization theory?"
Organization Studies, vol 13/1, pp. 1-17.
Parsons, T. (1960) Structure and process in modern societies, Glencoe, Ills: The Free
Press.
Pauka, T. and R. Zunderdorp (1988) De banaan wordt bespreekbaar, Cultuurverandering
216
in ambtelijk and politiek Groningen (transí. Discussing the banana, cultural change
within governmental and political Groningen), Amsterdam: Nijgh & van Ditmar.
Pedler, M., J. Burgoyne, and T. Boydell (1991) The learning company, McGraw Hill Book
Company.
Pennings, J.M. (1992) "Structural contingency theory: a reappraisal" Research in
Organizational Behavior, Vol 14. pp 267-309.
Pennings, J.M. and F. Harianto (1992) The diffusion of technological innovation in the
commercial banking industry, Strategic Management Journal, Vol 13, pp. 29-46.
Pentland, B.T. (1992) "Organizing Moves in software support hot lines" Administrative
Science Quarterly, vol 37, pp. 527-548.
Perez, C. (1985) "Microelectronics, long waves and world structural change: new
perspectives for developing countries", World Development, vol 13/3, pp. 441-463.
Perrow, C. (1986) Complex organizations: a critical essay, New York: Random House.
Peters, T.J. (1992) Liberation Management: necessary disorganization for the nanosecond
nineties, New York: Alfred A Knopf.
Peters, T.J. and R.H. Waterman (1982) In Search of excellence, London: Harper and Row.
Pettigrew, A.M. (1985) "Contextual research and the study of organizational change
processes" In: E. Mumford et al (ed) Research methods in information systems,
Amsterdam: Elsevier Science.
Pfeffer, J. and G.R. Salancik (1978) The external control of organizations: a resource
dependence perspective, New York: Harper & Row.
Pinch, T.J. and W.E. Bijker (1984) "The social construction of facts and artefacts: or how
the sociology of science and technology might benefit each other", Social studies of
science, vol 14, pp. 399-441.
Polanyi, M. (1958) The tacit dimension, New York: Anchor.
Porter, M.E. (1980) Competitive strategy: Techniques for analyzing industries and
competitors, New York: The Free Press.
Porter, M.E. (1990) The competitive advantage of nations, London: MacMillan.
Powell, W.W. (1991) "Expanding the scope of institutional analysis". In: W.W. Powell and
P.J. DiMaggio The new institutionalism in organizational analysis, Chicago:
University of Chicago Press.
Powell, W.W. and P.J. DiMaggio (ed) (1991) The new institutionalism in organizational
111
analysis Chicago: University of Chicago Press.
Powell, W.W., K.W. Koput, and L. Smith-Doerr (1996) "Interorganizational collaboration
and the locus of innovation: networks of learning in biotechnology", Administrative
Science Quarterly, vol 41/1, pp. 116-145.
Power, M. (1990) "Modernism, postmodernism and organisation". In: J. Hassard and D.
Pym (ed) The theory and philosophy of organizations, Routledge, Kegan and Paul.
pp. 109-124.
- Prahalad, C.K. and G. Hamel (1990) "The core competence of the corporation", Harvard
Business Review, may-june, pp. 79-91.
Putnam, L.L. and M.E. Pacanowsky (1983) Communication and organizations, An
interpretive approach, Newbury Park, CA: Sage Publications.
Quinn, J.B. (1989) "Strategic change: logical incrementalism" Sloan Management Review,
Vol 45, pp 45-60.
Raesfeld Meijer, A.M. Von, K. de Ruyter, P. Cabo (1996) "Corporation in new service
development: Social dynamic approach" In: TT.A. Swartz, D.E. Bowen, and S.W.
Brown, Advances in service marketing and management, JAI Press, vol. 5, pp. 193214.
Rogers, C.R. (1951) Client-centered therapy: Its current practice, implications, and theory,
Boston: Houghton Mifflin.
Rogers, E.M. (1983) Diffusion of innovations, third edition, New York: The Free Press.
Roth, G.L. and P.M. Senge (1996) "From theory to practice: research territory, processes
and structure at an organizational learning center", Journal of Organizational
Change Management, vol 9/1, pp. 92-106.
Rorty, R. (1979) Philosophy and the mirror of nature, Princeton: Princeton University
Press.
Rosenberg, N. (1982) Inside the black box, Cambridge: Cambridge University Press.
Rosenberg, N. and C. Frischtak (ed) (1985) International technological transfer, New york:
Praeger.
Ryles, G. (1949) The concept of mind, London: Hutchinson.
Sahal, D. (1981) Patterns of technological innovations, Reading, MA: Addison-Wesley.
Sahlin Anderson, K.
(1991) Imitating and editing success, The construction of
organizational fields and identities, paper presented at the 10th EGOS colloquim in
Wien.
218
Salaway, G. (1987) "An organizational learning approach to information systems develop­
ment" MIS Quarterly, June, pp 245-264.
Sandelands, L.E. and R. Drazin (1989) On the language of organization theory,
Organization Studies, vol 10, pp. 457-478.
Sandelands L.E. and R.E. Stablein (1987) "The concept of organization mind". In: S.
Bacharach and N. DiTomaso (ed) Research in the Sociology of Organizations, vol 5
Greenwich, CT: JAI Press, pp. 135-161.
Schein, E.H. (1992) Organizational culture and leadership 2nd edition, San Fransisco:
Jossey-Bass.
Schön, D.A. (1963) "Champions for radical new inventions" Harvard Business Review,
March-April, pp. 77-86.
Schön, D.A. (1967) Technology and change, New York: Delacorte.
Schön, D.A. (1982) The reflective practioner, New York: Basic Books.
Schön, D.A. (1971) Beyond the stable state, New York: Norton Library.
Schumpeter, J. (1934) The theory of economic development, Cambridge, MA: Harvard
University Press.
Schutz, A. (1971) Collected papers, vol 1 and 2, The Hague: Nijhoff.
Scott, W.R. (1987) "The adolescene of institutional theory", Administrative Science
Quarterly vol 32, pp. 493-511.
Scott, W.R. (1990) "Technology and Structure: an organizational-level perspective". In: P
S Goodman and L. Sproull, Technology and organizations, Jossey-Bass: San
Francisco, pp. 109-143.
Scott, W.R. (1992) Organizations: rational, natural and open systems, Englewood Cliffs,
N.J.: Prentice Hall.
Scott, W.R. (1983) "The organization of environments: network, cultural, and historical
elements", In: J.W. Meyer & W.R. Scott, Organizational environments: ritual and
rationality, Beverly Hills, CA: Sage.
Selznick, P. (1957) Leadership in administration, New York, Harper & Row.
Senge, P. (1990) "The leaders new work: Building learning organizations", Sloan
Management Review, fall, pp. 7-22.
Senge, P. (1992) The fifth discipline, the art & practice of the learning organization,
Randon House London.
Shibutani, T. (1955) "Reference groups as perspectives", American Journal of Sociology,
219
vol. 60, pp. 562-569.
Shrivastava, P. (1981) Strategic decision making process: the influence of organizational
learning and experience PhD dissertation, University of Pittsburg.
Shrivastava, P. (1^^) "A typology of organizational learning systems", Journal of
Management Studies, vol 20/2, pp. 7-28.
Silverman, D. (1970) The theory of organizations, London: Heinemann Ecucational Books
Simon, H.A. (1977) The new science of management decision, Englewood cliffs, NJ:
Prentince Hall.
Simon, H.A. (1985) "What we know about the creative process", In: T.S. Kuhn (ed)
Frontiers in creative and innovative management, Cambridge, MA: Balinger, pp. 320.
Sims, D. (1996) "organizational learning as the development of a canon of stories",
Proceedings of the Symposium "Organizational learning and the learning or­
ganization, Theoretical and research developments, Lancaster 1-3 September.
Sluijs van, E. (1991) De etnografie in het organisatie-onderzoek, MERIT 91-029,
University of Limburg.
Sminia, H. (1994) Turning the wheels of change, Groningen theses in Economics,
Management and organization, Worters-Noordhoff.
Smircich, L. and G. Morgan (1992) "The management of meaning", Journal of Applied
Behavioral Science, Vol 18/3, 257-273.
Stalk, G., P. Evans, L.E. Shulman (1992) "Competing on capabilities: the new rules of
corporate strategy", Harvard Business Review, vol. 70, pp. 57-69.
Stata, R. (1989) "Organizational learning - the key to management innovation" Sloan
Management Review, spring, pp. 63-74.
Stamper, R. (1992) "Signs, organizations, norms and information systems", Proceedings
Third Australian Conference on Information Systems, 5-8 October, pp. 21-54.
Starbuck, W.H. (1992) "Learning by knowledge-intensive firms", Journal of Management
Studies, vol 29/6, pp. 713-740.
Starbuck, W.H. (1983) "Organizations as action generators" American Sociological Review,
vol 48, pp. 91-102.
Stein, E.W. (1995) "Organizational memory: Review of concepts and recommendations for
management", International Journal of Information Management, vol 15/2,
pp. 17-32.
220
Stein, E.W. and V. Zwass (1995) "Actualizing organizational memory with information
systems" Information Systems Research, vol 6/2, pp. 85-117.
Strauss, A. (1978) Negotiations, San Francisco: Jossey-Bass.
Swieringa, J. and A.F.M. Wierdsma (1990) Op weg naar een lerende organisatie. Over het
leren en opleiden van organisaties (transí. Becoming a learning organization, On
learning and teaching in organizations), Groningen: Wolters Noordhoff.
Tenkasi, R.V. and R.J. Boland (1996) "Exploring knowledge diversity in knowledge
intensive firms: a new role for information systems", Journal of Organizational
Change, vol 9/1, pp. 79-91.
Thatchenkery, T.J. (1996) "Organizational learning, language games and knowledge
creation", Editorial note, Journal of Organizational Change Management, vol 9/1,
pp. 4-11.
Thomas, W.I. (1928) The child in American, behavior problems and programs, New York:
Knopf.
Thompson, J. (1967) Organizations in action New York: McGraw-Hill.
Tibosch, M.J.M.H. and M.S.H. Heng (1994) "Information systems and organizational
culture: a view point" Malaysian Journal of Management Science, vol. 3/2, pp. 1733.
Turban, E. (1995) Decision support systems and expert systems, fourth edition, Englewood
Cliffs, NJ: Prentice Hall.
Tushman, M. L. and T.J. Scanlan, (1981) "Boundary spanning individuals: Their role in
information transfer and their antecedents" Academy of Management Journal, vol
24, pp. 289-305.
Veld, i n ' t R.J., L. Schaap, C.J.A.M. Termeer and M.J.W. Twist (ed) (1991) Autopoiesis
and configuration theory: new approaches to societal steering. Dordrecht: Kluwer.
Vennix, J.A.M. (1990) Mental models and computer models, PhD thesis, Nijmegen.
Vissers G. (1994) The production of strategy, PhD thesis, Erasmus University Rotterdam,
Delft: Eberon.
Walsh, J.P. and G.R. Ungson (1991) "Organizational memory" Academy of Management
Review, vol 16/1, pp. 57-91.
Walsham, G. (1991) "Organizational Metaphors and information system research",
European Journal of Information Systems, vol 1/2, pp. 83-94.
Watson, W.E., K. Kumar and L.K. Michaelsen (1993) "Cultural diversity's impact on
221
interaction process and performance: comparing homogeneous and diverse task
groups" Academy of Management Journal, vol. 36, pp. 590-602.
Webber, A.M. (1993) "What's so new about the new economy?" Harvard Business Review,
january-february, pp. 24-42.
Weggeman, M. and T. Boekhoff (1995) Kenniswerkers en Kennismanagement (transí.
Knowledge workers and Knowledge management) Holland Management Review, vol
42, pp. 80-89.
Weick, K.E. (1990) "Technology as equivogue: sensemaking in new technologies", In: P.S.
Goodman and L. Sproull (ed), Technology and organizations, Jossey-Bass: San
Francisco.
Weick, K.E. (1979) The social psychology of organizing, Reading, MA: Addison-Wesly.
Westenholz, A. (1993) "Paradoxical thinking and change in the frames of references",
Organization Studies, vol 14/4, pp. 37-59.
Williamson, O.E. (1975) Markets and hierarchies, New York: Free Press
Williamson, O.E. (1985) The economic institutions of capitalism: firms, markets, relational
contracting, New York: Free Press.
Wolcott, H. (1975) "Criteria for an ethnographic approach to research in schools", Human
Organization, vol 34, pp. 111-128.
Woodward, J. (1958) Management and technology, London: H.M.S.O.
Woodward, J. (1965) Industrial organization, theory and practice, London: Oxford
University Press.
Wijnhoven, F. (1995) Organizational learning and information systems, The case of
monitoring information and control systems in machine bureaucratic organizations,
PhD thesis, University of Twente.
Yin, R.K. (1989) Case study research: design and methods, Applied social research method
servies, vol 5, Newsbury Park: Sage Publications.
Zee, van der H. (1996) "Kennismanagement als ontwikkelingsvraagstuk"
(transí.:
Knowledge management as a problem of development), In: J.W.M. Kessels and C
A. Smit (ed) Kennisproductiviteit (transí.: Knowledge productivity) Opleiders in
organisaties, Capita Selecta, vol 26, Deventer: Kluwer Bedrijfswetenschappen.
Zuboff, S. (1988) In the age of the smart machine, New York: Basic books.
Zucker, L.G. (1987) "Institutional theories of organizations" Annual Review of Sociology,
vol 13. pp. 443-464.
222
NEDERLANDSE SAMENVATTING
(SUMMARY IN DUTCH)
De aandacht voor 'organizational learning' neemt toe, zo blijkt uit de vele artikelen
en boeken die de laatste tijd over dit onderwerp verschijnen, en de conferenties en
workshops die eraan worden gewijd. Hoewel er blijkbaar behoefte bestaat het leren van
organisaties te bevorderen, heerst er over de betekenis van het begrip 'organizational
learning' nog veel onduidelijkheid. Dit kan onder andere worden afgeleid uit de inhoud
van deze boeken,
artikelen en congressen.
'Organizational
learning'
wordt op
uiteenlopende wijze benaderd. Het ontbreekt aan een gedeelde opvatting of benadering die
het begrip de benodigde theoretische onderbouwing zou verschaffen. Het gevaar dreigt dat
het concept binnen afzienbare tijd wordt bestempeld als het 'management buzzword' van
de jaren negentig.
Dit proefschrift is allereerst bedoeld om het concept te verduidelijken en
theoretisch te onderbouwen. Door middel van theoretische exploratie, aangevuld met
empirische inzichten, is getracht een wetenschappelijke bijdrage te leveren aan de
discussie over de betekenis van 'organizational learning'. De aandacht is daarbij
voornamelijk gericht op de dynamiek van organisatorische leerprocessen. Door leren als
een proces te beschouwen, komen problemen en complicaties aan het licht die het resultaat
van leren negatief kunnen beïnvloeden. Een procesbenadering maakt het mogelijk een
beter inzicht te verschaffen in de leercapaciteiten van organisaties. Dit is tevens de tweede
doelstelling van het proefschrift: gebaseerd op theoretische en empirische inzichten in de
wijze waarop organisaties leren, worden er uitspraken gedaan over het voorkómen van
leerproblemen en het bevorderen van leercapaciteiten.
Het proefschrift bestaat uit drie delen. In het eerste deel wordt er gezocht naar een
theorie over 'organizational learning'. Deze zoektocht leidt tot de conclusie dat de
literatuur zeer divers is en een aantal tekortkomingen kent. In reactie hierop wordt in deel
twee een alternatieve benadering van leren geïntroduceerd. In deel drie wordt op basis van
deze theoretische excursie geanalyseerd hoe organisaties hun leervermogen kunnen
bevorderen.
Deel I: Op zoek naar een theorie over organizational learning
Duidelijkheid verschaffen over de betekenis van het begrip 'organizational learning' vergt
allereerst de integratie van verschillende benaderingen ervan.
Hoofdstuk twee bevat een beschrijving van zes verschillende benaderingen van
'organizational learning'.
'Adaptief' leren, een van de oudste benaderingen, is voornamelijk gericht op
aanpassing aan de omgeving. In de 'increméntele innovatie'-benadering is leren het
diffusie-proces van externe kennis. Het perspectief dat is gericht op het 'uitwisselen van
aannames' berust op het uitgangspunt dat leren plaatsvindt als individuen hun persoonlijke
opvattingen aan elkaar kenbaar maken en de basis-aannames van de organisatie ter
discussie
stellen.
In
het
'organisatorische
kennis'-perspectief
is
leren
een
informatieverwerkend proces waarin organisatorische kennis wordt ontwikkeld. De
'lerende organisatie' heeft als perspectief een specifieke organisatievorm die positieve
resultaten van leren bevordert. Het 'sociaal-constructivistische'-perspectief is gericht op
informele leerprocessen die plaats vinden gedurende dagelijkse activiteiten.
Hoewel deze zes benaderingen elk hun eigen waardevolle bijdrage leveren, vormen
ze in combinatie met elkaar nog geen bevredigend alternatief. Alle benaderingen berusten
namelijk in meer of mindere mate op aannames die ertoe leiden dat alternatieven over het
hoofd worden gezien. In hoofdstuk drie worden vijf van zulke 'biases' besproken, alsmede
de mogelijkheden ze te vermijden.
Allereerst is er in de literatuur de neiging waarneembaar aan te nemen dat leren
per definitie resulteert in verbetering, intelligentie of vernieuwing. Dit vooruitgangsgeloof
is het resultaat van een uitkomst-benadering van leren. Leren kan echter ook resulteren in
stabiliteit, en zelfs in verslechtering of zelfdestructie. Om deze 'bias' te voorkómen, wordt
leren in dit proefschrift benaderd als proces, zodat de dynamiek van leren, inclusief de
mogelijke leerproblemen, aan de orde kan worden gesteld.
In de literatuur wordt leren veelal opgevat als individueel leren. Om leren op een I
hoger niveau van abstractie te brengen, wordt in dit proefschrift gebruik gemaakt van de ^
'institutionalisatie'-theorie van Berger en Luckman (1966).
^
In de literatuur overheerst het sy^emdeiiken als model om leermotieven te I
verklaren. Als gevolg hiervan werden andere motieven zoals experimenteren en de wens i
van individuen om zich te bewijzen of risico's te nemen, ten onrechte buiten beschouwing. [
Ook de factor toeval speelt in het systeemdenken geen rol van betekenis.
224
In de literatuur treft men voorts vaak de uitgangspunten aan dat organisaties
kunnen anticiperen op hun toekomst en dat leren radicale veranderingen tot stand brengt.
V Leren gebeurt echter ook vaak onopgemerkt of bij toeval. Daarnaast belemmeren de
geschiedenis en de bestaande kennis van de organisatie de mogelijkheden voor radicale
vernieuwingen. Anticiperen en 'doublé loop learning' zijn wel mogelijk, maar spelen in de
praktijk niet de doorslaggevende rol die sommigen veronderstellen.
^
1
*é ^ *
Ten slotte wordt leren wel gezien als een één- of tweezijdig fenomeen. Er is dan
sprake van slechts één aspect van leren, bijvoorbeeld het leren tijdens increméntele
innovaties, óf twee, tegengestelde processen waarbij één inferieur is. 'Single loop
learning' is bijvoorbeeld ondergeschikt aan 'doublé loop learning' (Argyris and Schön
1978) en 'adaptive learning' is minder waard dan 'generative learning' (Senge 1992). In
^dit proefschrift worden vier vormen van leren geïntroduceerd die alle, anders dan in de
i literatuur, wederzijds van elkaar afhankelijk zijn.
Deel II: Een typologie van 'organizational learning'-processen
Nadat in het eerste deel van dit proefschrift is aangegeven wat in de literatuur de
problemen zijn met 'organizational learning', wordt in deel twee een mogelijke oplossing
gepresenteerd. Daartoe worden de waardevolle bijdragen van de zes beschouwingen over
leren geïntegreerd, rekening houdend met de vijf theoretische condities zoals besproken in
hoofdstuk drie.
De hoofdstukken vier tot en met zeven behandelen de vier conceptueel
verschillende leervormen. Er is vanuit gegaan dat leren een evolutionair proces is waarin
valkuilen en obstakels kunnen vóórkomen die de uitkomst van leren sterk beïnvloeden. De
vier wijzen van leren die worden besproken zijn 'intern leren', 'feedback leren', 'leren
van anderen' en 'creatief leren'. De argumentaties berusten op een combinatie van
bestaande theoretische inzichten en op twee case studies. De eerste studie handelt over de
leerproblemen die een organisatie-afdeling 'Informatiesystemen' ondervond of juist niet
onderkende. De tweede studie gaat over het leerproces dat een groep 'innovators'
doormaakt tijdens de ontwikkeling van een product-idee. Beide studies dienen ter
illustratie van de theorie.
Intern leren is het basisproces waarin de organisatie leert van haar leden en de
leden van de organisatie. Het berust op de drie momenten van Berger en Luckman (1966):
225
externalisatie, objectificatie, internalisatie. Intern leren is, net als andere vormen van
leren, niet gevrijwaard van valkuilen en obstakels.
Van 'feedback leren' is sprake als de organisatie leert van feedback-informatie uit
de omgeving. Een belangrijk deel van de literatuur over 'organizational learning' is
gebaseerd op deze vorm van leren.
'Leren van anderen' betreft het leren van de ervaringen van andere organisaties.
Deze vorm van leren kan doelgericht zijn, bijvoorbeeld door middel van 'benchmarking'.
Vaak echter is er sprake van min of meer onbewust leerproces, zoals bij het inhuren van
nieuwe medewerkers of organisatie-adviseurs.
'Creatief leren' betreft het opdoen van nieuwe kennis door middel van
experimenteren, exploreren en creëren. De ideeën achter de 'lerende organisatie' zijn
voornamelijk gebaseerd op deze vorm van leren.
Deel III: Implicaties
Deel drie van het proefschrift behandelt de implicaties van wat tot zover hoofdzakelijk op
een conceptueel niveau is beschreven. In hoofdstuk acht wordt de vraag gesteld: "hoe
kunnen organisatorische leerprocessen resulteren in positieve uitkomsten zoals verbetering,
intelligentie en innovatie?". Gegeven het feit dat leren om verschillende redenen kan
worden bemoeilijkt, is de aandacht vooral gericht op de oorzaken van deze leerproblemen
en het vermijden ervan. Leerproblemen kunnen worden herleid tot twee fundamentele
oorzaken: 'focussed' leren en ongebalanceerd leren.
'Focussed' leren is het gevolg van een beperkte selectie en interpretatie van
informatie. De oorzaken hiervan zijn de macht van referentiegroepen, zelf-referentieel
informatiegebruik,
fysieke
en culturele condities,
specialisatie,
en het feit dat
leerprocessen vaak onopgemerkt blijven. 'Focussed' leren leidt vaak tot een 'pathdependent' ontwikkelingsproces. 'Focussed' leren kan worden voorkómen door een betere
communicatie en een hoger zelfbewustzijn. De macht van referentiegroepen en zelfreferentieel informatiegebruik vraagt daarnaast vaak om een 'derde actor', die in staat is te
interveniëren en, als buitenstaander, leerproblemen te analyseren.
Van ongebalanceerd leren is sprake in situaties waarin de organisatie teveel is
gericht op één van de vier leervormen. Intern leren bijvoorbeeld zou moeten worden
aangevuld met vormen van feedback leren, omdat anders onherroepelijk 'path-dependency'
optreedt en op de langere termijn zelfs zelfdestructie. Het balanceren van de vier
226
leervormen is niet alleen noodzakelijk om 'path-dependency' te vermijden, maar ook om
andere negatieve resultaten te voorkómen. Hoofdstuk acht bevat een beschouwing over de
wederzijdse afhankelijkheden van alle mogelijke combinaties van leervormen.
Organisaties kunnen ook gebruik maken van informatiesystemen als hulpmiddel om
hun leerprocessen te ondersteunen. In hoofdstuk negen van het proefschrift wordt de rol
van informatiesystemen tijdens leren geanalyseerd. Zes informatie-intensieve processen die
het leren kenmerken worden onderscheiden: externalisatie, objectificatie, internalisatie,
selectie, interpretatie en idee-generatie.
'
Externalisatie kan worden ondersteund met kennis-management systemen, zoals
Intranet en Lotus Notes. De discussie is voornamelijk gericht op de beperkingen van deze
systemen.
Informatiesystemen
die de objectificatie
van kennis ondersteunen,
worden
'Organizational Memory Information Systems' (OMIS) genoemd. Ook bij deze systemen
wordt een aantal kritische kanttekeningen geplaatst.
Gedurende het proces waarin organisatorische kennis wordt geïnternaliseerd, spelen
informatiesystemen een rol bij het verspreiden van basisaannames. Als gevolg hiervan
fungeren informatiesystemen vaak als systemen van manipulatie.
Ook het selecteren van kennis en informatie beïnvloedt het succes van leren. Zoals
is betoogd in hoofdstuk acht, zijn leerproblemen dikwijls het gevolg van 'focussed' leren.
In dit hoofdstuk wordt gesteld dat de meeste informatiesystemen deze wijze van leren
bevorderen en zo de problemen mede veroorzaken.
Het interpreteren van informatie kan ook het succes van leren beïnvloeden. Terwijl
verschillende interpretaties onduidelijkheden met zich kunnen brengen, kunnen gedeelde
interpretaties
leiden
tot
'path-dependency'.
Informatiesystemen,
zoals
Groupware
systemen, kunnen het uitwisselen van verschillende interpretaties mogelijk maken.
Informatiesystemen die het proces van idee-generatie ondersteunen, zijn niet alleen
van belang voor het bevorderen van creatief leren, ze kunnen ook nodig zijn ter
voorkoming van
'focussed'
leren.
De mogelijkheden
van simulatie-software
en
Groupware-systemen, de meest geaccepteerde informatiesystemen die idee-generatie
bevorderen, worden kort besproken. Onbeperkte exploratie vergt echter om andere
systemen. Als alternatief worden 'Alien Information Systems' (AIS) voorgesteld. AIS's
bestaan uit informatie waarvan de relevantie niet op voorhand kan worden vastgesteld,
227
waardoor toeval en serendipiteit een grotere rol kunnen spelen. Informatiesystemen die tot
deze categorie behoren zijn bijvoorbeeld bibliotheken, Intranet en Internet.
Het hoofdstuk mondt uit in de conclusie dat meer onderzoek nodig is naar de
mogelijkheden van informatiesystemen om 'organizational learning' te bevorderen.
228
The Tinbergen Institute is the Netherlands Research Institute and Graduate School for
Economics and Business, which was founded in 1987 by the Faculties of Economics and
Econometrics of the Erasmus University in Rotterdam, the University of Amsterdam and
the Free University in Amsterdam. The Institute is named after the late Professor Jan
Tinbergen, Dutch Nobel Prize laureate in economics in 1969.
The Tinbergen Institute is located in Amsterdam and Rotterdam.
If available trade editions of the books which are published in the Tinbergen Institute
Research Series can be ordered through Thesis Publishers, P.O. Box 14791, 1001 LG
Amsterdam, The Netherlands, phone: +3120 6255429; fax: +3120 6203395.
The following books recently appeared in this series:
85.
A.H.M. LELIVELD, Social security in developing countries; operation and
dynamics of social security mechanisms in rural Swaziland.
86.
Y.M. PRINCE, Price cost margins in Dutch manufacturing: with an emphasis on
cyclical and firm size effects.
87.
W.E. KUIPER, Farmers, prices and rational expectations.
88.
B. ROORDA, Global total least squares. A method for the construction of open
approximate models from vector time series.
89.
A.I. MARTINS BOTTO DE BARROS, Discrete and fractional programming
techniques for location models.
90.
J.A. DOS SANTOS GROMICHO, Quasiconvex optimization and location theory.
91.
R.B. KOOL, Aspects of enlarging the market effects in the Dutch health care.
92.
B. LEEFTINK, The desirability of currency unification theory and some evidence.
93.
A. VAN VLIET, Lower and upper bounds for on-line bin packing and scheduling
heuristics.
94.
C.C.J.M. HEYNEN, Essays in derivatives pricing theory.
95.
Y.M. VAN EVERDINGEN, Adoption and diffusion of the European, currency
unit. An empirical study among European companies.
96.
R.G. DE VILDER, Endogenous business cycles.
97.
J.J. STIBORA and A. DE VAAL, Services and services trade: A theoretical
inquiry.
98.
R. VAN DER BIE, "Een doorlopende groote roes". De economische ontwikkeling
van Nederland, 1913-1921.
99.
W.J. JANSEN, International capital mobility and asset demand. Six empirical
studies.
100.
N.E. STROEKER, Second-hand markets for consumer durables.
101.
J.E. VAN DEN BERG, Trade union growth and decline in The Netherlands.
102.
P.R.J. VAN DER LAAG, An analysis of refinement operators in inductive logic
programming.
103.
E. BEINAT, Multiattribute value functions for environmental management.
104.
H.A. VAN KLINK, Towards the borderless mainport Rotterdam. An analysis of
functional, special and administrative dynamics in port systems.
105.
W.H.J. HASSINK, Worker flows and the employment adjustment of firms. An
empirical investigation.
106.
S. LIU, Contributions to matrix calculus and applications in econometrics.
107.
C M . VAN PRAAG, Determinants of successful entrepreneurship.
108.
E. VERHOEF, Economic efficiency and social feasibility in the regulation of road
transport externalities.
109.
I. BRAMEZZA, The competitiveness of a European city and the role of urban
management in improving the city's performance. The cases of the Central Veneto
and Rotterdam regions.
110.
G.G.A. BIESSEN, East European foreign trade and system changes.
111.
A. LUCAS, Outlier robust unit root analysis.
112.
C M . FOKKEMA, Residential moving behaviour of the elderly: an explanatory
analysis for the Netherlands.
113.
T. EROL, Exchange rate policy in Turkey. External competitiveness and internal
stability studied through a macromodel.
114.
M.A. DE RUYTER VAN STEVENINCK, The impact of capital imports Argentina 1970-1989.
115.
D.R. DANNENBURG, Actuarial credibility models: evaluations and extensions.
116.
C. FOLKERTSMA, On equivalence scales.
117.
A. PLAAT, Research Re: search & Re-search.
118.
A.W. VAN DER KROGT, Service development in less developed countries. The
case of telecom services in Chile and Argentina.
119.
O.W. STEENBEEK, Financial regulation in Japan. Systemic risks and the Nikkei
futures market.
120.
T.R.P.J. KROES, Financial intermediation and monetary transmission.
121.
T.H.F. CHEUK, Exotic options.
122.
K. TEPLA, Accounting for the quality of work-life-putting price tags on sources of
job (dis)satisfaction.
123.
R.H.M. EMMERINK, Information and pricing in road transport: theoretical and
applied models.
124.
T.J.S. OFFERMAN, Beliefs and decision rules in public good games - theory and
experiments.
125.
F. MERCURIO, Claim pricing and hedging under market imperfections.
126.
A.W.M. ODE, Migrant workers in the Dutch labour market today.
111.
R.E. WILDEMAN, The art of grouping maintenance.
128.
M.T. ROCHA DE MAGALHAES MELO, Stochastic Lot-sizing in Production
Planning - Strategies for Make-to-Order and Make-to-Stock.
129.
G. RUSSO, Firms' recruitment behaviour. An empirical investigation of the use of
search channels, the rate of arrival of applicants, and the spatial radius of search.
130.
J.N. VAN OMMEREN, Commuting and relocation of jobs and residences; a
search perspective.
131.
F.N. GOUWELEEUW, A general approach for computing loss probabilities in
finite-buffer queues.
132.
M.R.H. MANDJES, Rare event analysis of communication networks.